<![CDATA[This study aims to develop and test a hand drawing classification model using the Support Vector Machine (SVM) algorithm to identify digits in American Sign Language (ASL). This method utilizes image processing techniques for extracting relevant features from hand images and SVM's ability to separate complex patterns. The training and test data consists of hand images representing the digits 0 through 9 in ASL. Tests are performed using test data that the model has never seen during training, to measure the performance and validity of the model in real-world situations. The results showed that the developed classification model was able to recognize digits in ASL with satisfactory accuracy, where the accuracy of the developed model was 99.8% with a loss of 0.018. Error analysis provides insight into situations that confuse the model and the potential for further improvement. The use of SVM in this ASL classification opens up new opportunities in strengthening communication accessibility for the hand sign language community. In conclusion, this model has the potential to make a positive contribution in facilitating communication and inclusion for deaf communities.]]>
Copyrights © 2023
