• Dhanish Kumar
• Nishant Kumar

As you can see, HTR in processer idea or photo handling is a significant and testing investigate area affecting different fields. It is useful for, amongst other things, examining bank payments, instructions and decoding characters on all kinds of methods. OCR (Recognition of Optical Characters ) Knowledge, customized for written by hand document translation, is one of the most important engines responsible for translating an entire character range from different file formats such as word or image documents. HTR faces challenges, including complex blueprint enterprises, diverse script styles, minor datasets, and inferior accurateness. Current progress in written by hand script acknowledgement (HTR) is largely due to deep learning and machine learning advances. This paper presents a hybrid method to tackle challenges in written by hand text recognition. The aim is to improve the accuracy of identifying handwritten text in images. Based on these observations, Deeplabcut2 presents the design of a neural network composed of a 5-layer. Convolutional Neuronal Network (CNN) linked to a Bidirectional Long Short- Term Memory (BiLSTM) forward-backward layer, surveyed by a Connectionist Sequential Organization (CTC) decoder. The proposed hybrid model accomplished 98.50 % and 98.80 % accurateness on the IAM and RIMES datasets, separately. This highlights the power and success of these neural network architectures, showing that applying them consecutively can increase accuracies in handwritten text recognition.

"Hybrid Deep Neural Network for Accurate Handwriting Recognition Using CNN-BiLSTM-CTC", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 4, page no.g615-g624, April-2025, Available :http://www.jetir.org/papers/JETIR2504675.pdf

"Hybrid Deep Neural Network for Accurate Handwriting Recognition Using CNN-BiLSTM-CTC", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 4, page no. ppg615-g624, April-2025, Available at : http://www.jetir.org/papers/JETIR2504675.pdf