Preprint / Version 1

Chess Board State Detection and Applications through Artificial Intelligence and Machine Learning, and Convolutional Neural Networks


  • Aditya Verma None



Chess, Artificial Intelligence, Machine Learning, Image Processing, Neural Networks, Computer Science, Recognition Systems, Software, Coding


This paper addresses the enhancement of automated chess recognition systems,
focusing on overcoming challenges such as poor lighting, diverse camera angles, and varying
chess set designs. Through the use of cutting-edge convolutional neural networks (CNNs)
trained on diverse datasets, this study employs advanced image processing techniques
including edge, corner, and line detection algorithms, along with the RANSAC algorithm for
robust corner identification. Pretrained CNN models are used for classifying board occupancy
and piece types. The methods significantly improve system accuracy under controlled
conditions, demonstrating high success rates in LED lighting on homogeneous surfaces.
However, performance is still affected by extreme lighting variations, unconventional chess sets,
non-standard camera angles, and glare reflected by the chessboard. The study presents
promising advancements, highlighting the potential for further improvements to enhance
universal applicability and robustness in real-world scenarios.


Acher, M., & Esnault, F. (2016). Large-scale analysis of chess games with chess engines: A preliminary report. arXiv preprint arXiv:1607.04186 [cs.AI]. Retrieved from

Angst, R. (2014, April 30). RANSAC: Random Sampling And Consensus [PowerPoint slides]. Stanford Electrical Engineering. Retrieved February 15, 2024, from

Berent, A. (2019, March 2). Forsyth-Edwards Notation. Adam Berent Software & Hobbies. Retrieved August 18, 2023, from

Bishop, C. M. (2013, February 13). Model-based machine learning. The Royal Society Publishing. Retrieved October 4, 2023, from

Chess board with chess set in opening position 2012. (2012, November 3). Wikimedia Commons. Retrieved April 15, 2024, from

Chess History: The Astonishing History of Chess (With Timeline). (2021, October 15). The Chess Journal. Retrieved January 3, 2024, from

Christensson, P. (2011, April 1). Grayscale. TechTerms. Retrieved March 7, 2024, from

Cicchetti, A., Ciccozzi, F., Mazzini, S., Puri, S., Panunzio, M., Zovi, A., & Vardanega, T. (2012). CHESS: A model-driven engineering tool environment for aiding the development of complex industrial systems. In Proceedings of the 27th IEEE/ACM International Conference on Automated Software Engineering (ASE '12) (pp. 362-365).

Gusev, D. A. (2021, June 16). Using Modern Chess Software for Opening Preparation. ERIC Institute of Educational Sciences. Retrieved July 24, 2023, from

Holt, J. (2024). 25 Unique and Unusual Chess Sets. Beautiful Life. Retrieved May 2, 2024, from

Khoche, K., Gurav, S., Pundir, R., Chrotiya, S., & Narooka, P. (2019). Application based smart chess board using interactive GUI design. International Journal of Computer Science Trends and Technology (IJCST), 7(1), 49. Retrieved from

Koray, C., & Sumer, E. (2016). A computer vision system for chess game tracking. In L. Cehovin, R. Mandeljc, & V. Štruc (Eds.), 21st Computer Vision Winter Workshop, Rimske Toplice, Slovenia, February 3–5, 2016 (pp. 1-7). Retrieved from

Li, Z., Liu, F., Yang, W., Peng, S., & Zhou, J. (2021). A survey of convolutional neural networks: Analysis, applications, and prospects. IEEE Transactions on Neural Networks and Learning Systems, 33(12), 6999-7019.

Mahesh, B. (2020). Machine learning algorithms - A review. International Journal of Science and Research (IJSR), 9(1). Retrieved from

Mendez, K. M., Pritchard, L., Reinke, S. N., & Broadhurst, D. I. (2019). Toward collaborative open data science in metabolomics using Jupyter Notebooks and cloud computing. Metabolomics, 15, Article 125.

Panchal, H., Mishra, S., & Shrivastava, V. (2021). Chess moves prediction using deep learning neural networks. In 2021 International Conference on Advances in Computing and Communications (ICACC) (pp. Page numbers). IEEE.

Wang, W., Tan-Torres, A., & Hamledari, H. (2017). Lecture #06: Edge Detection. Stanford Department of Computer Science. Retrieved November 20, 2023, from

Wei, Y.-A., Huang, T.-W., Chen, H.-T., & Liu, J.-C. (2017). Chess recognition from a single depth image. In 2017 IEEE International Conference on Multimedia and Expo (ICME) (pp. Page numbers). IEEE.

Wölflein, G., & Arandjelović, O. (2021). Determining chess game state from an image. Journal of Imaging, 7(6), 94.

Wölflein, G. (2021, May 25). Determining chess game state from an image. GitHub. Retrieved September 22, 2023, from

Wölflein, G., & Arandjelović, O. (2021, April 25). Dataset of Rendered Chess Game State Images. OSF. Retrieved August 3, 2023, from

Xie, Y., Tang, G., & Hoff, W. (2018). Chess piece recognition using oriented chamfer matching with a comparison to CNN. In 2018 IEEE Winter Conference on Applications of Computer Vision (WACV) (pp. Page numbers). IEEE.