• Riya
• Sorab Kumar

CAPTCHA provides way to secure the authentication process by differentiating humans from bots. CAPTCHA includes complex images and identifying characters from the image could be difficult task. This paper presents techniques including KNN, SVM and neural network-based approach commonly used for character recognition. Literature survey presents the detail of models used for character recognition. The results and limitations of KNN, SVM and neural network-based approach were highlighted to determine best possible approach out of these approaches. The results obtained from KNN, SVM and neural network-based approach are compared and SVM based approach yield best possible approach in terms of classification accuracy. The classification accuracy obtained from KNN, SVM and neural network-based approach was 85%, 92% and 90%. The compared techniques are best among character recognition mechanism, but further modifications cold be done in terms of pre-processing and feature extraction. In future, hybridization of pre-processing, optimized feature extraction along with SVM can be used to enhance the process of character recognition.

CAPTCHA, KNN, SVM , CNN

"Recognition Of Captcha Characters Using Machine Learning Algorithms", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.10, Issue 10, page no.b570-b573, October-2023, Available :http://www.jetir.org/papers/JETIR2310158.pdf

"Recognition Of Captcha Characters Using Machine Learning Algorithms", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.10, Issue 10, page no. ppb570-b573, October-2023, Available at : http://www.jetir.org/papers/JETIR2310158.pdf