After acquiring the images by the camera, the computing device then computes a plurality of moving traces MT of the objects O in the images in step S03. In detail, the camera acquires a first image at a first acquisition time, the computing device then uses a neural network deep learning method to identify the objects in the first image, and a plurality of first coordinates of the objects in the first image, wherein a confidence value of the identification result is higher than a threshold value. The camera then acquires a second image at a second acquisition time, the computing device then uses the neural network deep learning method to identify the objects in the second image, and a plurality of second coordinates of the objects in the second image, wherein a confidence value of the identification result is higher than a threshold value. The computing device then obtains the moving traces MT of every object in the images based on the first coordinate and the second coordinate of every object. In other words, the computing device computes the moving trace MT of each object based on the neural network deep learning method for the confidence value and coordinates of each object. The computing device is, for example, a central processing device or a cloud server with computing function of a monitoring center; and the neural network deep learning method is, for example, convolutional neural network (CNN) of the artificial intelligence (AI) technology. The present disclosure is not limited thereto.