Low Illumination Color Image Enhancement Algorithm Based On Retinex Theory

Image enhancement technology is widely used in several applications such as in the fields of public safety,industrial production and aerospace.In the recent years so studying image enhancement is of great significance.When acquiring digital images,the brightness and contrast of the images are reduced due to the effects of weather,light intensity,and shooting angle.Currently,image enhancement algorithms based on Retinex theory have significantly enhances the brightness of the image,and can realize local recognition details.However,there are problems such as loss of color information,low saturation,and severe local halo.Based on Retinex theory,combined with bilateral filtering and HSV color space model,this thesis proposes an enhancement algorithm for low-illumination color images.The algorithm uses bilateral filtering to perform global weighted averaging on the image,which can better protect the detailed edge information of the image and reduce the halo while denoising.After the light component that should be removed is processed and then combined with the reflection component,After the image is processed,it will be clearer and protect the color information.Then the improved algorithm is applied to the luminance component in the HSV space model,compared with the result calculated by the traditional algorithm in the RGB space model,the algorithm can better protect the color information in the image and reduce the color distortion,and reduce the amount of calculation.In this chapter,the subjective and objective evaluations of the image before and after processing by the enhancement algorithm are given respectively.It can be seen from the intuitive visual effects and evaluation index data that the algorithm proposed in this subject is superior to the existing traditional algorithms.