| Summary: | 碩士 === 國立勤益科技大學 === 電機工程系 === 98 === With the advancement in information technology, the information security and identity identification were increasingly important, using physiological characteristics to establish identity recognition system was the popular research project, so this paper used captures the thermal image of hand by using infrared camera. The extraction features include the length of Palm midpoint to each finger, hand profile, finger length and finger width as recognition features. Early identification system that captured the fingerprints or palmprints had to consider the light intensity, because the light intensity would affect the results of identification, so the processing of image enhancement and image filtering were used. But the processed images were lost the important features of information. Thermal images were sensing the infrared radiation of a specific band signal, the infrared signal can convert into image type for humans to observe, so there is no light interference problems, it still can get the feature data in the dark.
In conclusion, the extracted features were tested by using Extension method, extensive genetic algorithms and extension neural network, the experimental results confirmed that the proposed recognition system has a very high recognition rates, therefore, this paper using thermography of hand to identity recognition was feasible.
|
|---|
