Sign Language is a better language used by the hearing impaired. This report offers a means of visual communication over a very-low bandwidth communication network (i.e. telephone lines) of which the purpose is to supply a visual telecommunication among the deaf community. We develop a system which consists of the edge detection, the binarization, the vectorization, the inter-frame correspondence, the component trajectory finding, and hybrid encoding. This system is able to transmit a sequence of images of the sign language at higher resolution and lower transmission rate. I INTRODUCTIONThe hard-of-hearing population of Taiwan consists of approximately 200,000 people, of whom 20,000 are severely deaf. The population is further subdivided into people who are born deaf and those who become deaf. To communicate among themselves and with hearing people they utilize two means of expression : lips reading and sign language. Through history, the deaf have evolved manual sign languages to communicate with each other. Because of no widely accepted written form of any language has evolved, these languages have not developed quite the range of vocabulary and sophistication of the most advanced spoken language. Sign languages enable the deaf to communicate among each other with great facility, in contrast to the difficulty with which the deaf communicate with the hearing community by means of reading lips and facial expressions, and by means of written message. This paper presents an approach to develop a system that has been multinational interest in the transmission of moving pictures at very low rate (4.8k-19.2k bits/s).Tartter and Knowlton[1] have carried outexperiments at Bell Laboratories on communication between the deaf people using 27 luminous spots placed strategically on hands (with special glove) and nose. Their technique is encouraging but not very comfortable and requires a certain training time. Sperling[2] was the first one trying to solve the American Sign Language (ASL) compression problem. In his latter paper[6], he mention that for dynamic gray-level images of isolated ASL signs, reduction of the frame rate to 15 frames per second does not appreciably impair intelligibility. Thus the minimum bandwidth estimates immediately can be reduced fourfold to 5k Hz. Pearson and Six [3,4] have experienced the experiments with videophone over a period of 6 years.They have their investigation in binary images and outline cartoons. Their best results were with images of 50x50 pixels, with two-level intensity quantization, and with 12.5 frames per second. They mentioned that 100k bits/sec seems to be a threshold between the difficult communication; below 5k bits/s communication becomes impossible.Cohen et.al.[5,6,7J develop a compact hierarchical method of representation of binary images. The basic idea is to segment a picture into the largest possible uniform area and transmit a hierarchical representation of this area. They improved their previous biquad method and developed a vectorgraph coding method to code the lin...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.