Recognition of American Sign Language Using Modified Deep Residual CNN with Modified Canny Edge Segmentation

Abstract

American Sign Language (ASL) recognition system aims to recognise hand gestures’ meaningful motions, and it is a crucial solution to communicate between the deaf community and hearing people. However, existing sign language recognising algorithms still have some drawbacks, such as the difficulty of recognising hand movements low recognition accuracy for most of the sign language recognition. To address this problem, a Modified Convolutional Neural Network (MCNN) deep residual 101 classifier-based American Sign Language identification system is developed. There are three main parts present in the method. The first part is preprocessing the images to remove the noise, enhance the contrast of the picture, adjust the contrast level and smoothen the picture using various filters. The second part is the segmentation, and it’s used to partition the image using a modified canny edge detection method by removing all weak edges present in the image. Finally, classification will be done using the Modified CNN deep residual 101 classifiers. By classifying the image, the American Sign Language is accurately identified. This process is conducted through images. The outcome shows that the suggested approach has a 0.95 per cent accuracy and a 0.05 per cent False Positive Rate. Other CNN such as resNet 50 and resNet 18 reached 0.90% and 0.80% accuracy, respectively, which is lower than our proposed method. In addition, the resNet 101 classifier effectively recognises the difficult hand gestures through the image data and obtain high recognition accuracy for 36 signs is 0 to 9 numbers and a to z alphabets from American Sign Language.