Out-of-store Object Detection Based on Deep Learning

In the field of urban management, out-of-store operation is one of the key governance objects. Although there are many monitoring probes and large amounts video data, the management process is difficult due to the traditional technology used and the low efficiency of evidence collection. The concept of "Smart Urban Management" has introduced technologies such as mobile internet and cloud computing to realize the transformation of urban management into intelligent management. This paper proposed an out-of-store detection method, which combines image processing technology with deep learning model. The Faster R-CNN model is used to detect store locations and identify the out-of-store objects, and Visual Background Extractor (ViBe) method is applied to determine whether there is object outside of the store or not. Finally, a certain data processing method is used to record and collect evidence of the out-of-store operation phenomenon. The method is verified on the test data and the results show that it has a good detection effect which also prove its application value.