(Reuters) - An eight-year-old became the youngest player to defeat a grandmaster in classical chess on Sunday when Ashwath Kaushik beat Poland&#x27;s Jacek Stopa at the Burgdorfer Stadthaus Open in Switzerland.

Other Sports: Chess-Eight-year-old becomes youngest to beat grandmaster | The Star

The German teams continue to dominate the Mitropa Cup in Apolda, Germany. In round 6 both German teams won against France, and with this win Germany defended its lead in the Open while the women&#x27;s team took the sole lead. | Photos: German Chess Federation/Frank Hoppe

Mitropa Cup: Both German teams in the lead | ChessBase

The 1st Dortmund International Chess Days will take place from 10 to 18 August 2024 at the Cultural Centre Alte Schmiede in Dortmund. Frederik Sane and Leon Mendonca have already confirmed their participation in the A-Open. Elisabeth Pähtz and Dinara Wagner will play a match to find out who ...

Sparkassen Chess Trophy 2024 | ChessBase

Ashwath Kaushik of Singapore defeats 37-year-old grandmaster, beating record set days before by another eight-year-old

Eight-year-old becomes youngest chess player to beat grandmaster | Chess | The Guardian

At eight years, six months and 11 days old, Ashwath Kaushik made history by becoming the youngest player to beat a chess grandmaster in a classical tournament game.

Ashwath Kaushik: 8-year-old prodigy makes history after beating chess grandmaster | CNN

Senate Majority Leader Chuck Schumer (D-N.Y.) and 22 other Democratic senators sent a letter to Federal Trade Commission (FTC) Chair Lina Khan Wednesday demanding an investigation of two major proposed oil industry ... major U.S. energy companies dropped ...

Schumer leads 23 senators in demanding FTC probe of oil industry mergers

CEO John Hess is publicly enthusiastic about the deal, and it would be tough for another bidder to emerge given Hess’s size.

Chevron-Hess Deal on Track Despite Chevron Stock Drop

ExxonMobil's decision to buy Pioneer is a bet on the Permian Basin. On March 5, 2019, ExxonMobil revised its growth plans to produce more than 1 million barrels of oil equivalent per day (boe/d) from the Permian as early as 2024. ExxonMobil was on pace to miss that goal.

ExxonMobil and Chevron Are Betting Big on Fossil Fuels. Should You?

Milestone Environmental Services said it planned to lease about 22,000 acres in Midland and Loving counties in Texas.

Permian Basin energy companies focusing on carbon capture during region's growth

Exxon Mobil and Chevron told investors why they want to spend $114 billion combined on two megadeals. Their earnings reports revealed why they need to

Exxon and Chevron megadeals reveal why days of easy oil are over

Hess

hess

The Product and Technology team at The Markup has been busy with what we’ve lovingly called “Blacklight Summer,” and we’ve made some major updates to Blacklight, our popular real-time website privacy inspector. Blacklight popular As of August 18, Blacklight has three new features that allow you to: Scan sites from the EU: By default, Blacklight scans from a U.S. IP address. Now you can use the new options box to scan from Europe (the Amazon Web Services infrastructure where the scan originates is located in Hesse, Germany, to be precise).
View results from a desktop browser: Blacklight has always scanned from the perspective of a mobile browser. But now with the options box, you can choose to see results from a desktop view. This includes desktop screenshots in the Learn More section of scan results.
Scan specific URLs within a site’s structure: In the past, Blacklight could only scan a site’s homepage and one randomly selected page in the site’s structure. Now, if you enter a full URL for Blacklight to scan (e.g., https://themarkup.org/about), Blacklight will analyze just that specific page. Scan sites from the EU: By default, Blacklight scans from a U.S. IP address. Now you can use the new options box to scan from Europe (the Amazon Web Services infrastructure where the scan originates is located in Hesse, Germany, to be precise). Scan sites from the EU: View results from a desktop browser: Blacklight has always scanned from the perspective of a mobile browser. But now with the options box, you can choose to see results from a desktop view. This includes desktop screenshots in the Learn More section of scan results. View results from a desktop browser: Scan specific URLs within a site’s structure: In the past, Blacklight could only scan a site’s homepage and one randomly selected page in the site’s structure. Now, if you enter a full URL for Blacklight to scan (e.g., https://themarkup.org/about), Blacklight will analyze just that specific page. Scan specific URLs within a site’s structure: https://themarkup.org/about We built these new features because, well, they’re what users told us would be most useful. We hope that proves true. Specifically, we hope the new bells and whistles will help people in the following ways: told us would be most useful They’ll allow comparisons between how sites behave for visitors from the EU versus other regions.
Scanning individual pages within a site will show if certain pages have different trackers than other pages.
Similarly, being able to look at sites in both desktop and mobile view will illuminate whether sites use different sets of trackers, depending on how people access the sites. They’ll allow comparisons between how sites behave for visitors from the EU versus other regions. Scanning individual pages within a site will show if certain pages have different trackers than other pages. Similarly, being able to look at sites in both desktop and mobile view will illuminate whether sites use different sets of trackers, depending on how people access the sites. In addition to the new features above, we’ve been working diligently to reduce the number of timeout and error messages on certain sites. Errors may still occasionally occur, but users should encounter them much less frequently. If you have questions, see anything that looks not quite right, or just want to give the team your thanks, please email us at blacklight@themarkup.org. blacklight@themarkup.org Credits: Ramsey Isler Director of Product Ramsey Isler Illustration Gabriel Hongsdusit Gabriel Hongsdusit Gabriel Hongsdusit Engagement Maria Puertas Maria Puertas Maria Puertas Copy Editing and Production Sabrina Toppa Sabrina Toppa Sabrina Toppa Editors Michael Reilly Michael Reilly Michael Reilly Also published here Also published here here Photo by Bhushan Sadani on Unsplash Photo by Bhushan Sadani on Unsplash Bhushan Sadani Unsplash

The Blacklight Privacy Inspector Is Gaining Three New Features 

What Makes a Scene Important? This AI Knows

Author:
(1) Tiffany N. Younger, The Graduate Center, City University of New York, New York, NY 10016, USA (tyounger@hunter.cuny.edu). Author: Author: (1) Tiffany N. Younger, The Graduate Center, City University of New York, New York, NY 10016, USA (tyounger@hunter.cuny.edu). Table of Links Abstract and 1 Introduction Abstract and 1 Introduction 2. Early Life 2. Early Life 3. Whiteness as an Institution 3. Whiteness as an Institution 4. The Triple Threat 4. The Triple Threat 5. The Academic Plantation Field 5. The Academic Plantation Field 6. The Future Is Black Women 6. The Future Is Black Women 7. Imagination as a Tool 7. Imagination as a Tool 8. Imagination through Research 8. Imagination through Research 9. Imagination through Play 9. Imagination through Play 10. Collaboration as Imagination 10. Collaboration as Imagination 11. Conclusions and References 11. Conclusions and References 7. Imagination as a Tool Whiteness operates as a diversion, serving as a dehumanizing strategy. Gendered antiblack racism is a form of psychological terrorism. One way to disrupt both whiteness and gendered anti-black racism without being manipulated is by exercising our imagination. We can be creators of new epistemologies that center our own experiences and aspirations. Our collective consciousness and imagination are necessary tools to build a free world. Our purpose as Black women scholars should not be confined to solely deconstructing and discrediting white epistemologies. We must situate our purpose and career by envisioning our own ontological futures. Below are three examples of how I plan to use imagination in my work within academia moving forward. 8. Imagination through Research As I enter my role as an early career researcher, I enter ready to center Black women’s lived experiences in all my work. I believe Black women’s narrative is essential information for pushing toward liberation and overall economic justice. Black women are unique; they make up 70% of primary/sole heads of households in the United States (Price et al. 2020). Yet, most of our work on economic disparities and health disparities is studied through a race lens. These race-centered methods center on a cis-hetero patriarchal approach and perspective in research and interventions. This approach lacks substance and true centering on the lived experiences and needs of Black households and communities. A true racial justice approach includes gender at the center. I plan to use my time as a scholar working with Black women across the country to build a data bank of our lived experiences and desires. I plan to use my imagination to create new research methodologies that center joy, imagination, and creativity. Too often, Black women are overly researched in what we call “trauma porn” for the sake of data which results in more money for the institution. Furthermore, they are rarely compensated for their worth and value in participating and are left with a USD 25 gift card and a broken heart. My desire is for my research to be a space of refuge for Black women to come and just BE. 9. Imagination through Play Somehow, in the United States, an individual’s career becomes their identity. I know this because I have traveled and worked in other parts of the world, such as South Africa, Brazil, Ghana, Kenya, and the Dominican Republic, and work rarely comes up outside of the work environment. Gray et al. (2023) note, “Black women are more than mules of white supremacy”. It is essential to have hobbies and passions that spiritually sustain you and can be poured into you outside of work. I have recently been intentional about defining myself and exploring who I am outside of academia and my career as a research scientist and professor, as Black women have been conditioned to labor and serve through capitalism. Laboring and serving are not natural, despite what we may think. This is the conditioning we inherited from our ancestors, who were conditioned by white colonizers to labor excessively without rest or payment, which is another tenet in the savagery of whiteness. We must disrupt this conditioning and engage in nonpayment/production hobbies. Our imagination can be activated by drawing, playing mystery games with friends, singing karaoke, and playing solo solitaire. This summer, I plan to teach my daughter and niece how to jump double Dutch and hopscotch to reclaim my inner child. It was through play and engaging in joy that my imagination allowed me to BE. 10. Collaboration as Imagination I always carry an “ideas” notebook with me to write down ideas, and on the first page, there is a quote by Tricia Hershey that reads, “I do nothing alone, my whole life is a collaboration”. This quote grounds me and calls me to collaborate instead of isolate, which is my natural tendency. Collaboration is key in the work that we conduct; finding “our people” is necessary to keep us grounded and sane. Gems 2 and 4, outlined above in the critical reflection, expose the need for having people whom you can call when your vision is cloudy and your ability to imagine has been eclipsed by whiteness. Gem #2 Black women are my barometer, which has forced me to redefine my ideas of what success looks like. Just given the fact that Black women’s ideas are often viewed as DEI, it is already setting us back. I realized that if I remained true to myself, my community, and Black women-centered research, then I would risk not being successful in academia by getting awards, grants, or even tenure. I had a mentor, who is a Black woman, tell me, “Tenure can’t be the goal. It can’t define you”. She was telling me that I was more than a label of tenured professor and that is a white-centered standard that does not center my desires and the people I serve. So, I made Black women and other women of color my barometer. Anytime I am challenged institutionally, I collaborate with my circle on how to address the issues. Before any major hearing, or if someone “tells” on me and I am summoned by university administration and lawyers, I strategize with at least two other Black women, from adjunct faculty to Vice Provost, on how to deal with an issue. When I address my challenge, I repeat something my wise healing coach told me, “There is nothing for me to protect and defend”. I always feel armed and ready when I enter these complicated, treacherous institutional issues because at least two brilliant Black women have helped prepare me for the matter, and good luck trying to outsmart all of us. “I do nothing alone, my whole life is a collaboration” 11. Conclusions Black women, although to these institutions we are deemed a broken policy, we are not illegal or broken. Academic and social service institutions are a manifestation of power through the creation of ideologies of dominance such as whiteness, cis-heteropatriarchy, and coloniality. Our mere existence and being is a manifestation of liberation our ancestors got free; therefore, in every instance, there is hope. This manifesto is a call to create and not to advocate for the abandonment of resistance. We can challenge systems and structures of power and confront whiteness while simultaneously using imagination to shape our identities and desires. This manifesto is a call to center our humanity in our endeavors, recognizing that embracing our own ontology and epistemology is affirming our humanity Funding: This research received no external funding. Funding Conflicts of Interest: The author declares no conflict of interest. Conflicts of Interest: References Andrews, Kehinde. 2016. The psychosis of whiteness: The celluloid hallucinations of “Amazing Grace” and “Belle”. Journal of Black Studies 47: 435–53. Available online: http://www.jstor.org/stable/43926967 (accessed on 10 February 2024). [CrossRef] Brown, Tyese A. 2023. A Blackgirl Artivisionary Mosaic: Art-Based Participatory Refusals to School Punishment. Doctoral dissertation, City University of New York, New York, NY, USA. Burton, E. Thomaseo, Rebecca C. Kamody, Emily I. Pluhar, Emily Gray, and Syidah Abdullah. 2020. Radical Acceptance and Obesity-Related Health Conditions: A Case Report. Journal of Clinical Psychology in Medical Settings 27: 217–25. [CrossRef] [PubMed] Collins, Patricia Hill. 2000. Black Feminist Thought: Knowledge, Consciousness, and the Politics of Empowerment. New York and London: Routledge. Cooper, Brittney. 2014. Does anyone care about Black women? Meridians 12: 153–55. [CrossRef] Cooper, Brittney. 2018. Eloquent Rage: A Black Feminist Discovers Her Superpower. New York: St. Martin’s Press. Crenshaw, Kimberlé L. 1991. Mapping the margins: Intersectionality, identity politics, and violence against women of color. Stanford Law Review 43: 1241–99. [CrossRef] Gray, Aysa, Norka Blackman-Richards, and Selena T. Rodgers. 2023. The DEI Industrial Complex: Undermining Black Woman Leadership. In The Experiences of Black Women Diversity Practitioners in Historically White Institutions. Edited by Tristen Brenaé Johnson. Hershey: IGI Global, pp. 1–20. [CrossRef] Hartman, Saidiya. 2016. The belly of the world: A note on Black women’s labors. Souls 18: 166–73. [CrossRef] Hess, Barnor. 2007. Racialized modernity: An analytics of White mythologies. Ethnic and Racial Studies 30: 643–63. [CrossRef] Hooks, Bell. 1994. Teaching to Transgress. New York and London: Routledge, Taylor & Francis Group. Hooks, Bell. 2000. All About Love: New Visions. New York: William Morrow. Lorde, Audre. 1982. Learning from the 60’s [Speech]. Retrieved from Blackpast. Available online: https://www.blackpast.org/africanamerican-history/1982-audre-lorde-learning-60s/ (accessed on 10 February 2024). Melendez, Diana, Diana Ballesteros, Cameron Rasmussen, and Alexis Jemal. 2023. Igniting pathways for land-based healing: Possibilities for institutional accountability. Genealogy 7: 62. [CrossRef] Pon, Gordon. 2009. Cultural Competency as New Racism: An Ontology of Forgetting. Journal of Progressive Human Services 20: 59–71. [CrossRef] Price, Anne, Jhumpa Bhattacharya, and Kendra Bozarth. 2020. Mississippi Is America: How Racism and Sexism Sustain a Two-Tiered Labor Market in the U.S. and Constrict the Economic Power of Workers in Mississippi and Beyond. Englewood: Insight Center. Available online: https://insightcced.org/wp-content/uploads/2020/10/INSIGHT_Mississippi-Is-America-brief_3.pdf (accessed on 10 February 2024). Rodgers, Selena T. 2017. Womanism and Afrocentricity: Understanding the intersection. Journal of Human Behavior in the Social Environment 27: 36–47. [CrossRef] Rodgers, Selena T. 2021. Next wave of post traumatic slave syndrome survivors black women resisters in academia. Advances in Social Work 21: 438–59. [CrossRef] Spencer, Zoe, and Olivia N. Perlow. 2018. Sassy mouths, unfettered spirits, and the Neo-lynching of Korryn Gaines and Sandra Bland: Conceptualizing Post Traumatic Master Syndrome and the familiar “policing” of Black women’s resistance in twenty-first century America. Meridians: Feminism, Race, Transnationalism 17: 163–83. [CrossRef] Spillers, Hortense. 2013. Mama’s baby, papa’s maybe: An American grammar book. In Black, White, and in Color: Essays on American Literature and Culture. Chicago: University of Chicago Press, pp. 203–29. Spivak, Gayatri Chakravorty. 1988. Can the subaltern speak? In Marxism and the Interpretation of Culture. Edited by Cary Nelson and Lawrence Grossberg. Urbana: University of Illinois Press, pp. 21–78. Thomas, Kecia M., Juanita Johnson-Bailey, Rosemary E. Phelps, Ny Mia Tran, and Lindsay N. Johnson. 2013. Women of color at midcareer: Going from pet to threat. In The Psychological Health of Women of Color: Intersections, Challenges, and Opportunities. London: Bloomsbury Publishing, pp. 275–86. Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. Disclaimer/Publisher’s Note: This paper is available on arxiv under CC BY 4.0 DEED license. This paper is available on arxiv under CC BY 4.0 DEED license. available on arxiv

From Survival to Imagination: A Black Feminist Blueprint for Liberation

Author:
(1) Tiffany N. Younger, The Graduate Center, City University of New York, New York, NY 10016, USA (tyounger@hunter.cuny.edu). Author: Author: (1) Tiffany N. Younger, The Graduate Center, City University of New York, New York, NY 10016, USA (tyounger@hunter.cuny.edu). Table of Links Abstract and 1 Introduction Abstract and 1 Introduction 2. Early Life 2. Early Life 3. Whiteness as an Institution 3. Whiteness as an Institution 4. The Triple Threat 4. The Triple Threat 5. The Academic Plantation Field 5. The Academic Plantation Field 6. The Future Is Black Women 6. The Future Is Black Women 7. Imagination as a Tool 7. Imagination as a Tool 8. Imagination through Research 8. Imagination through Research 9. Imagination through Play 9. Imagination through Play 10. Collaboration as Imagination 10. Collaboration as Imagination 11. Conclusions and References 11. Conclusions and References 3. Whiteness as an Institution “Cultures of domination rely on the cultivation of fear as a way to ensure obedience”—Bell Hooks (2000) “Cultures of domination rely on the cultivation of fear as a way to ensure obedience”—Bell Hooks (2000) “Cultures of domination rely on the cultivation of fear as a way to ensure obedience”—Bell Hooks (2000) “Cultures of domination rely on the cultivation of fear as a way to ensure obedience” Before entering academia, I spent the first decade of my adulthood working in various social service organizations for a series of nonprofits and advocacy think tanks. Although the organizations were different in various ways, what was consistent across institutions was the organizational structure, expectations, and implicit punishment and silencing of those who failed to work within the confinements of the ideologies of dominance such as whiteness, coloniality, and patriarchy. Melendez et al. (2023) posit, “Ideologies of dominance at the root of racial-settler colonialism are embedded in all United States institutions”. Each institution I worked at had a commitment to social justice and liberation for all, yet they all perpetuated inequality to maintain power through the enforcement of policies designed to protect dominance through the centering of whiteness. Hess (2007) refers to the functioning of whiteness as “white mythologies” and argues that it functions as the dissonance between reality and what is socially fictitiously constructed. Race is not real; it is constructed. Whiteness is not real; it is also constructed. Somehow, there is a lack of interrogation of whiteness as a construct that ultimately positions whiteness as what is natural and normal, which results in its ability to dominate. Whiteness was created to establish the binary we see today through the formulation of the hierarchy of power, where the top of the hierarchy represents power and privilege, and the bottom of the hierarchy encompasses anti-blackness, serving as the foundation of exploitation. At the end of the hierarchy framework, the binary constructs of structural and systematic inequality functions. On one end of the spectrum, there is whiteness, which has power, and the other end is anti-blackness, which is the anchor of oppression. It is through these two constructs that inequality was designed and currently serves as the root of the United States economy. The economy we have today is a result of racial–settler colonialism and the money that allows these organizations to operate cannot be separated from it. Therefore, the root of the money institutions invest to start their work is at least two times stolen: capital from the stolen land from Native Indigenous people and stolen labor of Black enslaved people. My understanding of the history of institutional investment is what I bring to every position I have worked in. I operate as if all executives and managers who carry out inequality work are clear about the intentional design of our economy and inequality. I worked in the fields of social welfare and social services. Understanding the construction of the US economy is pertinent to the social welfare field. My experience as a young professional during my first decade of work in nonprofits and think tanks was a form of psychological abuse. In my experience, whiteness showed up in ways that made me question my reality. For example, from 2017 to 2020, I served as a board member at a small nonprofit and raised concerns that leadership did not reflect the community being served. I decided to resign after three years of waiting for promises of equity to be fulfilled. I was promised the first year of my tenure as a board member that the current director of the organization, who happened to be a white woman who was not from the community but had a Ph.D. in social work, would not become the Executive Director of the organization. I was promised by the founder, who was also a white woman, that the board would hire someone from the community to lead the organization. Three years later, the director became the Executive Director, and the justification was “she deserves the job after all of her hard work”. Whiteness builds the game and then dictates the rules of engagement so the outcome can be in its favor. Upon my resignation, one of the founders, who at the time I thought was a friend, responded to my resignation and stated, “I honestly had the vibe that you didn’t want this to work, and this was the outcome hoped for which you can use as a story in your career”. This statement was meant to question my intentions of resignation; it almost got me to question my reality and decision until I could point to the material, tangible fact: the Executive Director did not represent the community as promised. Intention over impact is the result of whiteness. Today, in 2024, the Executive Director has not changed, and she now oversees over eight Black and Latino staff members; all of the children they serve are Black and Latino, and not a single community member has a job within the organization. Whiteness is literally profiting from a community’s harm and marginalization when whiteness is the original designer and benefactor of their hardship. This is modern-day colonialism. Almost seven years later, the leadership has remained the same. I made the right decision. Whiteness is entitled “she deserves the job after all of her hard work” “I honestly had the vibe that you didn’t want this to work, and this was the outcome hoped for which you can use as a story in your career” The cycle and control of whiteness continued as I climbed the career ladder. My tenure working at a national gender equity think tank as the Director of Policy and Advocacy was my entry into the treacherous world of knowledge production. Language is vital in my work as a social policy scholar, researcher, and practitioner. Acknowledging all parts of history so that future scholars and practitioners have what they need to build future scholarship is imperative. As an emerging scholar, I believe it is my duty to be honest in my scholarship on inequality. While that desire might seem evident, it is not practiced by all scholars and practitioners. There is a dominant and intentional narrative framing in both public and private scholarship that centers on whiteness. Whiteness gets to tell the story from its viewpoint. The control of language is one of the primary functions of whiteness. I experienced language control firsthand, guised as a “best practice” skill while working at a national think tank. While drafting a report on the racial and gender wealth gap, I was told I could not use the term “anti-black” in my writing. When I asked the reason, I was told, “the language is too strong” and “it goes against our communication best practice”. My director at the time was a white woman with work experience in “women’s” wealth work (not gender wealth work) and a master’s degree in political economy. At the time, I was in my fourth year of doctoral studies, with two master’s degrees and over fifteen years of work experience. I am also a Black woman who has studied critical race scholarship for a decade. I was hired to work at the national level to center race and gender in the wealth work at the institution. As an emerging Black feminist scholar, I understood that words matter. I advocated for the term and asked for a description of our policies and definitions of “strong language”. I argued that there is a clear distinction between racism and anti-black racism. My director at the time told me I was “new” to this landscape, and she advised me to listen more before I tried to perform “new things”. As a Black woman completing my terminal doctoral degree with over fifteen years of formal higher education, I was not deemed expert enough by my director to use terminology I considered appropriate or lead our narrative work without her approval and comfort of language being used. This is just one example among many of how whiteness sustains itself by spreading falsehoods to shield its own existence. Gem #3. Do not listen to their words. Let actions and outcomes serve as information when whiteness tries to play in your face. Gem #3. Do not listen to their words. Let actions and outcomes serve as information when whiteness tries to play in your face. Gem #3. Do not listen to their words. Let actions and outcomes serve as information when whiteness tries to play in your face. Gem #3. Do not listen to their words. Let actions and outcomes serve as information when whiteness tries to play in your face. This paper is available on arxiv under CC BY 4.0 DEED license. This paper is available on arxiv under CC BY 4.0 DEED license. available on arxiv

Who Controls the Narrative? When Whiteness Dictates Social Change

Authors:
(1) VINÍCIUS YU OKUBO, Dept. Electronic Systems Engineering, Polytechnic School, University of São Paulo, Brazil;
(2) KOTARO SHIMIZU, Department of Applied Physics, The University of Tokyo, Tokyo 113-8656, Japan;
(3) B. S. SHIVARAM, Department of Physics, University of Virginia, Charlottesville, Virginia 22904, USA;
(4) HAE YONG KIM, Dept. Electronic Systems Engineering, Polytechnic School, University of São Paulo, Brazil. Authors: Authors: (1) VINÍCIUS YU OKUBO, Dept. Electronic Systems Engineering, Polytechnic School, University of São Paulo, Brazil; (2) KOTARO SHIMIZU, Department of Applied Physics, The University of Tokyo, Tokyo 113-8656, Japan; (3) B. S. SHIVARAM, Department of Physics, University of Virginia, Charlottesville, Virginia 22904, USA; (4) HAE YONG KIM, Dept. Electronic Systems Engineering, Polytechnic School, University of São Paulo, Brazil. Table of Links Abstract and I. Introduction Abstract and I. Introduction II. Related Work II. Related Work III. Methodology III. Methodology IV. Experiments and Results IV. Experiments and Results V. Conclusion and References V. Conclusion and References IV. EXPERIMENTS AND RESULTS A. PERFORMANCE EVALUATION A. PERFORMANCE EVALUATION A. PERFORMANCE EVALUATION Evaluating TM-CNN poses a certain challenge because there are no benchmarks or published results for direct comparison. So, we defined that junction and terminal detections reviewed by human observers as the gold standard and compared the algorithms against this standard. For testing purposes, we annotated 5 independent images from varying experiments, using the same procedure as the training images (subsection III-E). False detections were excluded since they do not contribute to the detection evaluation. In total, the test set contained 3,551 junctions and 3,653 terminals. B. RESULTS FROM A PHYSICS PERSPECTIVE B. RESULTS FROM A PHYSICS PERSPECTIVE B. RESULTS FROM A PHYSICS PERSPECTIVE In Bi:YIG films, it is known that a perfect stripe pattern is the most energetically favorable and stable configuration of magnetic moments [4]. In this structure, the magnetic moments exhibit stripes aligning straight in one direction throughout the entire structure, resulting in large spatial coherence. Meanwhile, the experimentally obtained structures often manifest labyrinthine patterns, as illustrated by dark and bright stripes in Fig. 1. In these labyrinthine patterns, stripes are aligned within small domains but propagate in different directions between different domains, and thus the spatial coherence of stripes is smaller than the perfect stripe. Notably, junctions and terminals are defects emerging in the formation of these labyrinthine patterns, intricately linked to the growth of spatial coherence in the system. Introducing such defects into the perfect stripe structure leads to the bending and branching of the initially aligned stripes, eventually reducing spatial coherence accompanied by the transformation to labyrinthine patterns. Indeed, a plethora of defects are observed in the labyrinthine patterns shown in Fig. 2. Conversely, the elimination of these defects plays a crucial role in the transition from a quenched state to an annealed state, where the system increases the spatial coherence. Each of these defects cannot be eliminated alone by continuous deformation of the stripes due to their topological properties. They can be removed through the pair annihilation. For instance, in Fig. 2, terminals are branched from junctions and they can be removed pair wisely by reducing the branching distance. These defects, known as topological defects, are widely recognized for playing a crucial role in the formation of periodic structures. By detecting the positions of defects, the evolution of the labyrinthine patterns would be systematically quantified. Specifically in the context of magnetic labyrinthine patterns, a much smaller dataset was used to detect these defects manually [4]. Recently, persistent homology has been used to extract the topological features of the labyrinthine patterns in a systematic way [38]. However, it is still difficult to detect the positions of defects and classify them into junctions and terminals. In this article, we achieved the systematic detection. To quantify the evolution of the labyrinthine patterns, we applied the TM-CNN technique to the 444 images and investigated how the number of defects and their locations changes step by step. Fig. 6 illustrates the step dependence of the number of junctions and terminals during the demagnetization process. Results from 6 different experimental runs were averaged to estimate the number of defects. In the initial stages of the demagnetization process, there are approximately 800 defects. The number of defects sharply decreases around step 5 and remains nearly unchanged from step 10 onward. Consequently, the transition to the quenched state is inferred to start before step 5 and be completed by step 10. The reduction in the number of defects during the demagnetization process is compatible with the naive expectation based on the physical argument of a decrease in the number of defects associated with the increased spatial coherence of the stripes. Furthermore, the number of junctions and terminals are close, which is consistent with the topological argument that junctions and terminals are paired. Further quantification of the evolution of the labyrinthine patterns based on the spatial distribution of the defects (and not only on their number) is discussed in another article [6]. V. CONCLUSION In this work, we presented a new algorithm named TMCNN to detect defects in magnetic labyrinthine patterns, contributing to a pioneering analysis in material science. Our study characterized the evolution of junctions and terminals in magnetic stripes during demagnetization procedures, aiming at better understanding defect arrangement in magnetic materials [6]. TM-CNN employs a two-stage detection procedure, combining template matching for initial detection and a convolutional network classifier for refining misdetections. This approach ensures a high detection accuracy and facilitates dataset annotation through a semi-automatic procedure. In our experiments, TM-CNN exhibited performance superior to other techniques, achieving an impressive F1 score of 0.988. This high performance is mainly due to TMCNN’s ability to locate small and clustered objects. TM-CNN achieves almost 100% accuracy with a simple CNN classifier with less than half a million parameters and can be used even on computers without GPUs. While TM-CNN was developed for defect detection in labyrinthine magnetic patterns, its potential applications are not limited to this field. Future research could explore the use of TM-CNN in other domains, such as identifying bifurcations in blood vessels or adapting it to other structures that can be modeled using templates. REFERENCES [1] Masaya Uchida, Yoshinori Onose, Yoshio Matsui, and Yoshinori Tokura. Real-space observation of helical spin order. Science, 311(5759):359–361, 2006. [2] T. Garel and S. Doniach. Phase transitions with spontaneous modulation the dipolar ising ferromagnet. Phys. Rev. B, 26:325–329, Jul 1982. [3] P. Molho, J. L. Porteseil, Y. Souche, J. Gouzerh, and J. C. S. Levy. Irreversible evolution in the topology of magnetic domains (invited). Journal of Applied Physics, 61(8):4188–4193, 04 1987. [4] M. Seul, L. R. Monar, L. O’Gorman, and R. Wolfe. Morphology and local structure in labyrinthine stripe domain phase. Science, 254(5038):1616– 1618, 1991. [5] Naoto Nagaosa and Yoshinori Tokura. Topological properties and dynamics of magnetic skyrmions. Nature Nanotechnology, 8(12):899–911, Dec 2013. [6] Kotaro Shimizu, Vinicius Yu Okubo, Rose Knight, Ziyuan Wang, Joseph Burton, Hae Yong Kim, Gia-Wei Chern, and B. S. Shivaram. Machine Learning Assisted Characterization of Labyrinthine Pattern Transitions. arXiv:2311.10558, 2023. [7] Kai Briechle and Uwe D. Hanebeck. Template matching using fast normalized cross correlation. In David P. Casasent and Tien-Hsin Chao, editors, Optical Pattern Recognition XII, volume 4387, pages 95 – 102. International Society for Optics and Photonics, SPIE, 2001. [8] P. Viola and M. Jones. Rapid object detection using a boosted cascade of simple features. In Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001, volume 1, pages I–I, 2001. [9] N. Dalal and B. Triggs. Histograms of oriented gradients for human detection. In 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’05), volume 1, pages 886–893 vol. 1, 2005. [10] Alex Krizhevsky, Ilya Sutskever, and Geoffrey E Hinton. Imagenet classification with deep convolutional neural networks. 25, 2012. [11] Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition, 2015. [12] Ross Girshick, Jeff Donahue, Trevor Darrell, and Jitendra Malik. Rich feature hierarchies for accurate object detection and semantic segmentation, 2014. [13] M. Everingham, L. Van Gool, C. K. I. Williams, J. Winn, and A. Zisserman. The pascal visual object classes (voc) challenge. International Journal of Computer Vision, 88(2):303–338, June 2010. [14] Shaoqing Ren, Kaiming He, Ross Girshick, and Jian Sun. Faster r-cnn: Towards real-time object detection with region proposal networks, 2016. [15] Joseph Redmon, Santosh Divvala, Ross Girshick, and Ali Farhadi. You only look once: Unified, real-time object detection, 2016. [16] Yang Liu, Peng Sun, Nickolas Wergeles, and Yi Shang. A survey and performance evaluation of deep learning methods for small object detection. Expert Systems with Applications, 172:114602, 2021. [17] Ricardo H Maruta, Hae Yong Kim, Danilo R Huanca, and Walter J Salcedo. A new correlation-based granulometry algorithm with application in characterizing porous silicon nanomaterials. ECS Transactions, 31(1):273, 2010. [18] Hae Yong Kim, Ricardo Hitoshi Maruta, Danilo Roque Huanca, and Walter Jaimes Salcedo. Correlation-based multi-shape granulometry with application in porous silicon nanomaterial characterization. Journal of Porous Materials, 20:375–385, 2013. [19] Sidnei Alves De Araújo, Jorge Henrique Pessota, and Hae Yong Kim. Beans quality inspection using correlation-based granulometry. Engineering Applications of Artificial Intelligence, 40:84–94, 2015. [20] Gui-Song Xia, Julie Delon, and Yann Gousseau. Accurate junction detection and characterization in natural images. International Journal of Computer Vision, 106(1):31–56, Jan 2014. [21] Harry Pratt, Bryan M. Williams, Jae Yee Ku, Charles Vas, Emma McCann, Baidaa Al-Bander, Yitian Zhao, Frans Coenen, and Yalin Zheng. Automatic detection and distinction of retinal vessel bifurcations and crossings in colour fundus photography. Journal of Imaging, 4(1), 2018. [22] Sheng He, Marco Wiering, and Lambert Schomaker. Junction detection in handwritten documents and its application to writer identification. Pattern Recognition, 48(12):4036–4048, 2015. [23] M Elena Martinez-Perez, Alun D Hughes, Alice V Stanton, Simon A Thom, Neil Chapman, Anil A Bharath, and Kim H Parker. Retinal vascular tree morphology: a semi-automatic quantification. IEEE Trans Biomed Eng, 49(8):912–917, August 2002. [24] K. Liu, Y.S. Huang, and C.Y. Suen. Identification of fork points on the skeletons of handwritten chinese characters. IEEE Transactions on Pattern Analysis and Machine Intelligence, 21(10):1095–1100, 1999. [25] Chungnan Lee and Bohom Wu. A chinese-character-stroke-extraction algorithm based on contour information. Pattern Recognition, 31(6):651– 663, 1998. [26] Michael Maire, Pablo Arbelaez, Charless Fowlkes, and Jitendra Malik. Using contours to detect and localize junctions in natural images. In 2008 IEEE Conference on Computer Vision and Pattern Recognition, pages 1–8, 2008. [27] Ran Su, Changming Sun, and Tuan D. Pham. Junction detection for linear structures based on hessian, correlation and shape information. Pattern Recognition, 45(10):3695–3706, 2012. [28] R. Deriche and T. Blaszka. Recovering and characterizing image features using an efficient model based approach. In Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, pages 530–535, 1993. [29] He Zhao, Yun Sun, and Huiqi Li. Retinal vascular junction detection and classification via deep neural networks. Computer Methods and Programs in Biomedicine, 183:105096, 2020. [30] Kaiming He, Georgia Gkioxari, Piotr Dollár, and Ross B. Girshick. Mask R-CNN. CoRR, abs/1703.06870, 2017. [31] I B Puchalska, G A Jones, and H Jouve. A new aspect on the observation of domain structure in garnet epilayers. J. Phys. D: Appl. Phys., 11(15):L175, oct 1978. [32] Hae Yong Kim and Sidnei Alves de Araújo. Grayscale template-matching invariant to rotation, scale, translation, brightness and contrast. In Domingo Mery and Luis Rueda, editors, Advances in Image and Video Technology, pages 100–113, Berlin, Heidelberg, 2007. Springer Berlin Heidelberg. [33] J. P. Lewis. Fast normalized cross-correlation. Vision Interface, 95:120, 1995. [34] Sidnei Alves de Araújo and Hae Yong Kim. Ciratefi: An rst-invariant template matching with extension to color images. Integr. Comput.-Aided Eng., 18(1):75–90, jan 2011. [35] Hae Yong Kim. Rotation-discriminating template matching based on fourier coefficients of radial projections with robustness to scaling and partial occlusion. Pattern Recognition, 43(3):859–872, 2010. [36] Li Deng. The mnist database of handwritten digit images for machine learning research [best of the web]. IEEE Signal Processing Magazine, 29(6):141–142, 2012. [37] Christian Eggert, Stephan Brehm, Anton Winschel, Dan Zecha, and Rainer Lienhart. A closer look: Small object detection in faster r-cnn. In 2017 IEEE International Conference on Multimedia and Expo (ICME), pages 421–426, 2017. [38] T. Yamada, Y. Suzuki, C. Mitsumata, K. Ono, T. Ueno, I. Obayashi, Y. Hiraoka, and M. Kotsugi. Visualization of topological defect in labyrinth magnetic domain by using persistent homology. Vacuum and Surface Science, 62(3):153–160, 3 2019. This paper is available on arxiv under CC BY 4.0 DEED license. This paper is available on arxiv under CC BY 4.0 DEED license. available on arxiv

What Can Defects Tell Us About Magnetic Materials? TM-CNN's Defect Whisperer Approach

Authors:
(1) VINÍCIUS YU OKUBO, Dept. Electronic Systems Engineering, Polytechnic School, University of São Paulo, Brazil;
(2) KOTARO SHIMIZU, Department of Applied Physics, The University of Tokyo, Tokyo 113-8656, Japan;
(3) B. S. SHIVARAM, Department of Physics, University of Virginia, Charlottesville, Virginia 22904, USA;
(4) HAE YONG KIM, Dept. Electronic Systems Engineering, Polytechnic School, University of São Paulo, Brazil. Authors: Authors: (1) VINÍCIUS YU OKUBO, Dept. Electronic Systems Engineering, Polytechnic School, University of São Paulo, Brazil; (2) KOTARO SHIMIZU, Department of Applied Physics, The University of Tokyo, Tokyo 113-8656, Japan; (3) B. S. SHIVARAM, Department of Physics, University of Virginia, Charlottesville, Virginia 22904, USA; (4) HAE YONG KIM, Dept. Electronic Systems Engineering, Polytechnic School, University of São Paulo, Brazil. Table of Links Abstract and I. Introduction Abstract and I. Introduction II. Related Work II. Related Work III. Methodology III. Methodology IV. Experiments and Results IV. Experiments and Results V. Conclusion and References V. Conclusion and References II. RELATED WORKS A. JUNCTIONS AND TERMINALS DETECTION A. JUNCTIONS AND TERMINALS DETECTION A. JUNCTIONS AND TERMINALS DETECTION Junctions and terminals are shapes with relevance that extend beyond materials. Within computer vision, recognizing and enumerating them has been performed in diverse contexts, such as natural landscapes [20], biology [21] and handwriting images [22]. B. CLASSICAL METHODS FOR JUNCTIONS AND TERMINALS DETECTION B. CLASSICAL METHODS FOR JUNCTIONS AND TERMINALS DETECTION B. CLASSICAL METHODS FOR JUNCTIONS AND TERMINALS DETECTION One prevalent approach for detecting junctions and terminals involves using skeletonization as a pre-processing step. This process reduces the image to one-pixel-wide lines to represent its structures. Points in skeleton can then be identified as terminals, junctions and crossings based on their neighboring pixels. This technique has been applied in vascular images [23] and in the analysis of handwritten Chinese characters [24]. Pre-processing techniques using contour information have also been explored for junction detection. Lee and Wu [25] investigated stroke extraction in Chinese characters. Their method segments regions according to their contour, identifying junctions by counting neighboring regions. Maire et al. [26] proposed a junction detector in natural images by locating intersecting contours. This approach applies an expectation–maximization style algorithm to iteratively select relevant contours and suggest the junction’s position. However, skeletonization and contour finding are noise sensitive processes, and a pre-processing error will lead to a detection error. Junctions and terminals can also be identified by analyzing the arrangement of linear structures within the image. Su et al. [27] describe a technique for identifying these linear structures using the Hessian Matrix. This approach has been validated in biological images such as blood vessels, neutrites and tree branches. Xia et al. [20] present a junction detection method in natural images, based on amplitudes and phases of the normalized gradients of the image. Template-based approaches quantify the similarity of the appearance of image regions and the template. Deriche and Blaszka [28] modeled this approach as energy minimization, which is calculated by the deviation between the image and a predetermined model. This enabled the detection of key image features, such as edges, corners and terminals. C. DEEP LEARNING METHODS FOR JUNCTIONS AND TERMINALS DETECTION C. DEEP LEARNING METHODS FOR JUNCTIONS AND TERMINALS DETECTION C. DEEP LEARNING METHODS FOR JUNCTIONS AND TERMINALS DETECTION Owing to the success of R-CNN based detection techniques, Pratt et al. [21] developed a pipeline for identifying junctions and crossings in retinal vascular structures. Their method involves two main steps: initially, detection regions are proposed centred along the blood vessels, which are then classified as junctions, crossings or background. To generate the detection regions, their approach requires a binary segmented version of the exam. Theses images undego a skeletonization process, with the resulting points serving as references for the centers of the blood vessels. Zhao et al. [29], addressing the same problem, proposed using a Mask R-CNN base model [30] for region proposal. This strategy enables inference without the need for binary segmented version of the exam. However, during training, the segmented images are still used in the Mask R-CNN model to enhance its learning capabilities. This approach surpassed the performance of previous techniques in the detection of junctions and crossings in retinal vascular images. III. METHODOLOGY A. DATASET A. DATASET A. DATASET In this study, we used films of a ferromagnetic material of recognized technological importance, Bi:YIG, and obtained magnetic images of the labyrinthine patterns using a microscope with polarized light [31]. We specifically focused on the evolution of the labyrinthine patterns under the demagnetization field protocol described below. First, we prepared the sample in the fully saturated state by applying a sufficiently large magnetic field in +z direction, which is perpendicular to the films. In this state, magnetic moments in the Bi:YIG film are forced to point in the field direction. Next, we instantaneously dropped the field to zero and hold it for 10 seconds to get the image. Since the magnetic field is zero, magnetic moments can point upward and downward, resulting in the labyrinthine patterns shown in Fig. 1; the bright and dark regions represent the domains with opposite directions of magnetic moments. Such a process involving switching on and off the magnetic fields is considered half of the demagnetization step. In the remaining half step, a magnetic field was applied with a reduced amplitude and oriented in the opposite direction. We again captured the magnetic domain image after reducing the magnetic field to zero. By repeating these protocols up to 18 steps, we investigated the evolution of the labyrinthine patterns in the demagnetization process step by step. The amplitude of the magnetic field was exponentially reduced with each step. We conducted a series of demagnetization processes from the fully saturated state, repeating this cycle six times. Furthermore, we explored another situation, where the magnetic field was initially applied in the −z direction, and its direction was alternated step by step. Consequently, a total of 12 demagnetization processes were performed, yielding a collection of 444 domain images. All measurements reported here were performed at room temperature. The experimentally obtained images covered an area of 2 mm × 1.8 mm. The original high-resolution color images were converted to grayscale and their resolutions were reduced to 1300×972 to facilitate processing. Furthermore, a median filter with kernel size 3 was applied to reduce noise. B. TM-CNN OVERVIEW B. TM-CNN OVERVIEW B. TM-CNN OVERVIEW Our approach to detect junctions and terminals in magnetic labyrinthine patterns consists of two sequential steps: proposal of potential detections and their classification between junction, terminal and false detections. It is inspired by other cascaded object detection techniques like Viola and Jones face detection [8], R-CNN [12], and scale and rotation invariant template matching [32]. Fig. 3 illustrates the overall structure of TM-CNN. In the first phase, the algorithm generates a preliminary set of potential detections. It must propose all true defects, even if it also generates many false positives. We achieve this by applying template matching detection with a low threshold, followed by a non-maximum suppression. In the second phase, in order to eliminate the false positives, each potential detection is filtered by a CNN classifier. C. TEMPLATE MATCHING C. TEMPLATE MATCHING C. TEMPLATE MATCHING The basic form of template matching finds instances of a smaller template T within a larger image I. This is done by calculating some similarity metric between the model T and the content of a moving window located at each possible position of I. We measured the similarity between the image and the template using the Normalized Cross Correlation (NCC). NCC is invariant to linear changes in brightness and/or contrast. NCC between template T and image I at pixel (x, y) is calculated as: This basic approach is not well suited for detecting junctions and terminals in the magnetic labyrinthine structures because a single template is not capable of modeling: All possible rotations;


All deformed shapes of defects. All possible rotations; All possible rotations; All deformed shapes of defects. All deformed shapes of defects. To solve problem (1), we employ a rotation-invariant template matching based on exhaustive evaluation of rotated templates. There are some alternative rotation-invariant techniques based on circular and radial projections [32], [34], and on Fourier coefficients of circular and radial projections [35]. These techniques can reduce computational requirements, but their implementations are complex and require parameter tuning. Furthermore, our application does not require exceptional computational performance, as the processing is offline. Thus, our technique uses the standard OpenCV[1] template matching implementation, which is highly optimized using FFT and special processor instructions. D. TEMPLATES AND MASKS USED IN THE EXPERIMENT D. TEMPLATES AND MASKS USED IN THE EXPERIMENT D. TEMPLATES AND MASKS USED IN THE EXPERIMENT We manually designed templates and masks, and empirically tuned them to capture various junction and terminal shapes. They have 21×21 resolutions and are generated at runtime. Templates represent magnetic strips as black lines radiating from their center drawn on a white background. Meanwhile, masks are created from black backgrounds with white areas defining relevant coordinates used in template matching. Their main purpose is to obscure the space between strips and background, addressing variations in widths and curvatures, and to limit the background regions to reduce interference from neighboring strips. Table 1 exemplifies the templates and masks used to detect junctions and terminals. Template matching is applied separately for each (template, mask) pair. Several template matchings are computed in parallel using the OpenMP library[2]. This process takes about 80 seconds to process an image on an i7-9750H processor. To obtain the final correlation map corr, we calculate the maximum value among all n = 3 × 439 + 5 × 120 = 1917 NCC maps for each position (x, y), that is: Pixels where corr(x, y) exceeds a predefined threshold t are considered potential detections. However, a single junction/terminal may encompass multiple neighboring points with correlation values greater than the threshold t. Therefore, it is necessary to perform some form of non-maximum suppression to eliminate duplicate detections and select only the true center of junction/terminal. Kim et al. [18] present a solution to this problem: Whenever two potential detection points p1 and p2 are separated by a distance smaller than a threshold c, the point with the lowest correlation value is discarded. In this work, we use a slightly different approach, but with the same practical result: Whenever the algorithm finds a potential detection, it executes a breadth-first search algorithm. This algorithm recursively searches adjacent pixels (x, y) where the correlation exceeds 80% of the threshold (that is, corr(x, y) > 0.8t) and saves the pixel with the highest correlation. Subsequently, the searched area has its correlation value set to zero to avoid re-detection. Fig. 3d highlights the searched area in red. The pixel with the highest correlation is chosen as the center of the junction/termination. This process performs detection in a single pass. E. DATASET ANNOTATION E. DATASET ANNOTATION E. DATASET ANNOTATION To classify potential detections into true or false using a CNN classifier, we must first create annotated training images. TM-CNN makes it easier to create training examples, as it allows one to annotate the examples semi-automatically. This process is divided in two phases. 1) Template matching-assisted annotation 1) Template matching-assisted annotation In this phase, only a small set of images are annotated. Initially, we apply template matching followed by nonmaximum suppression to identify the centers of possible detections, together with their probable labels (junction or terminal). Without this help from template matching, we would have to manually and precisely locate the centers of thousands of defects. Subsequently, a human reviewer makes corrections to ensure that the labels given by the template matching are correct. The reviewer may change the labels to junction, terminal or false detection. After all positive detections are annotated along with a small set of false detections, a larger set of false detections is created by lowering the template matching threshold and sampling new false detections. These images, now with positive and negative annotations, are used to train a preliminary version of the CNN classifier. 2) Deep learning-assisted annotation 2) Deep learning-assisted annotation Due to the small number of images in the initial training set, the preliminary CNN classifier cannot accurately classify all magnetic stripe defects. Nonetheless, this preliminary model is integrated into the annotation procedure to alleviate the required workload. In the second phase, we continue using template matching to generate the initial set of detections. However, the preliminary CNN classifier is employed to identify most of the template matching errors, thus speeding up the annotation process. As new images are annotated, more accurate models are trained to further simplify the annotation workload. The final training set consists of 17 images derived from a single annealing protocol, selected to cover varied experimental configurations of ascending and descending magnetic fields at different magnitudes. Out of these, 16 were selected from the quenched (unordered) state, as they represent a more diverse set of shapes and represent a greater challenge for classification. The training set encompasses a total of 33,772 detections, which includes 12,144 junctions, 12,777 terminals, and 8,851 false detections. F. CANDIDATE FILTERING BY CNN F. CANDIDATE FILTERING BY CNN F. CANDIDATE FILTERING BY CNN Our algorithm extracts small 50×50 patches centered around each detection point and a CNN classifies them into three classes: junction, terminal or false positive. The size of patches for CNN classification is larger than the size of template matching models (21×21), allowing CNN to use more contextual information than template matching. We use a simple CNN model to classify the small patches (Fig. 4). It has four convolutional layers with 32, 64, 128 and 256 filters, all using 3×3 kernels. The first three convolutional layers are followed by max pooling layers to downsample the feature maps and global max pooling is applied after the last convolutional layer. This is followed by dropout and two fully connected layers: the first with 128 nodes and the second with three output nodes. The ReLU activation function is used across the model, except at the output layer where the softmax is used. In total, this network has only 422,608 parameters. For comparison, VGG-16 and ResNet-50 (common backbones for detection) have 138 million and 23 million parameters, respectively. Thanks to its simplicity, our model is fast and can make predictions even without GPUs and takes around 30 seconds for filtering each image using an i7-9750H processor with 16 GB of RAM. This paper is available on arxiv under CC BY 4.0 DEED license. This paper is available on arxiv under CC BY 4.0 DEED license. available on arxiv [1] Open Source Computer Vision Library, https://opencv.org [2] Open Multi-Processing, https://www.openmp.org/ [3] https://www.tensorflow.org/

Stories, News & Mentions about HESS

Categories

Trending Topics

Stories, News & Mentions about HESS

Hess’s Stories on HackerNoon

Hess's News & Mentions

Categories

Trending Topics