With the increasing use of digital video surveillance systems in the past few years, a greater number of images are being stored and analyzed. Thus, large amounts of time and labor are required to browse stored images when searching for specific people and objects. Based on the concept of video synopsis, we proposed an image processing flow to decrease the number of images requiring browsing, thus reducing the browsing time of surveillance images. We first used the discrete wavelet transform (DWT) to reduce the image resolution. We also used a low-pass filter to filter out the noise and fake motion generated from the external environment. Next, we removed the redundant information from the timeline, which allowed us to delete images in which no incidents occurred, then we used the proposed algorithm to extract the foreground objects. Finally, we calculated the space information of the recorded images, then a continuous event was integrated and merged in an image, thus reducing the time required for video playback.