Continuous Recognition Method for Japanese Finger Letters Using Contour Feature
| Autor: | Masahiro Kaneda, Shuichi Tokunaga, Hirokazu Osaki |
|---|---|
| Rok vydání: | 1996 |
| Předmět: |
Computer science
business.industry General Engineering Image processing Sign language Japanese Sign Language language.human_language Edge detection Image (mathematics) Numeral system Pattern recognition (psychology) Feature (machine learning) language Computer vision Artificial intelligence business |
| Zdroj: | JSME international journal. Ser. C, Dynamics, control, robotics, design and manufacturing. 39:78-84 |
| ISSN: | 1340-8062 |
| DOI: | 10.1299/jsmec1993.39.78 |
| Popis: | This paper deals with a continuous recognition method for finger signs used to express some numerals in Japanese sign language by image processing technology. There have been some reports about recognition methods for finger letters; however, these were concerned with the finger letters which could be recognized from only the hand shape. Some finger letters are expressed by both hand shape and motion; therefore, it is necessary to use sequential hand images to recognize such finger letters. We propose a new continuous recognition method for sequential hand images and their contour curves. First, we define some properties of the hand image with its contour curve. Next, we determine a judgement rule for discrimination of the hand shapes based on the above properties. Finally, we describe a recognition rule to apply the judgement rule to some sequential hand images, and show some examples to demonstrate the feasibility of the continuous recognition method. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|