Gray spot detection in surveillance video using convolutional neural network
Liang Hu, Li Chen, Jun Sheng Cheng · 2018
Video surveillance systems have been widely used in society and plays an important role in the maintenance of public security and social justice. Since the camera has been in the natural environment for a long time it is vulnerable to all kinds of interference resulting in the recorded video information is no longer of monitoring value. In this paperwe propose a detection method based on convolution neural network aiming at the problem of dust speckle interference in video surveillance. We train a fully convolutional network for segmentation and a convolutional neural network for classification simultaneously. Experimental results show that using the result after the classification as the input of the segmentation can reduce the false positives rate of the segmentation result. The experiment shows that our method has achieved good results. In the background of big datait has advantages over traditional algorithms.