Adaptive threshold vein enhancement method based on illuminance estimation
Autor: | Xinhua Liu, Hailan Kuang, Fuying Guan, Xiaolin Ma |
---|---|
Rok vydání: | 2021 |
Předmět: |
ComputingMethodologies_SIMULATIONANDMODELING
Computer science business.industry Quantitative Biology::Tissues and Organs Gaussian ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Illuminance 020206 networking & telecommunications 02 engineering and technology Filter (signal processing) Image (mathematics) Noise symbols.namesake Signal-to-noise ratio Computer Science::Computer Vision and Pattern Recognition Lab color space 0202 electrical engineering electronic engineering information engineering Image noise symbols 020201 artificial intelligence & image processing Computer vision Artificial intelligence business |
Zdroj: | 2021 IEEE 5th Advanced Information Technology, Electronic and Automation Control Conference (IAEAC). |
Popis: | Infrared vein images often contain a lot of noise due to the performance defects of the infrared imaging system, and these images have the characteristics of low contrast, low signal-to-noise ratio, and blurred target details. In order to improve this situation, an adaptive threshold enhancement algorithm based on guided filter illuminance estimation (ATEBIE) is proposed to enhance the detailed information of the vein infrared image in this paper, especially the vein pattern, which is of great significance for applications such as vein display and vein recognition. Firstly, the infrared image of the vein is converted to the LAB color space representation to obtain the L-component of the image; Secondly, the multi-scale Gaussian blurring is performed on the component, and the noise is removed at the same time to obtain the initial illuminance estimation; then the guided filtering is used for accuracy illumination estimation; Finally, according to the illuminance image, an adaptive threshold is adopted to dynamically adjust the image to achieve the purpose of vein image enhancement. The experimental results show that the proposed algorithm not only suppresses the image noise very well, but also effectively enhances the vein texture details, and has a good application prospect. |
Databáze: | OpenAIRE |
Externí odkaz: |