Nighttime low illumination image enhancement is highly desired for outdoor computer vision applications. However, few works have been studied towards this goal. In addition, the low illumination enhancement problem becomes very challenging when the depth information of a low illumination image is unknown. To address this problem, in this paper, we propose a dual channel prior-based method for nighttime low illumination image enhancement with a single image, which builds upon two existing image priors: dark channel prior and bright channel prior. We utilize the bright channel prior to get an initial transmission estimate and then use the dark channel as a complementary channel to correct potentially erroneous transmission estimates attained from the bright channel prior. Experimental results show significant credibility of the approach both visually and by quantitative comparison with existing methods.

Convolutional networks are powerful visual models that transform images into more effective representations. To make full use of this technique, we propose a new method based on deep learning and convolutional network to effectively get the discharge information of an ultraviolet (UV) image. We firstly segment the equipment region and the UV spot region separately by using the DeepLab network, and then several properties which can show the discharge information are extracted on the basis of the segmentation. The use of the DeepLab network helps to get a reliable segmentation result, and more accurate discharge information. We take 5000 UV images to test our network, and use the concept mean IOU to evaluate its performance. The results show the advantage of the method, and it can meet the demands of further fault diagnosis.