Efficient Abnormal Motion Change Detection Based on Translational Velocity Indicator

The translational velocity indicator(TVI) estimation is a simple and convenient way of solving for objects' motion direction without segmentation or tracking. Based on this, we defined and distinguished the change level of image sequences by defining Abnormality Coefficient (AC). An abrupt increase in AC can indicate the happening of abnormal motion change. To gauge the accuracy of judgment, we also introduced the width of S-band which is related to the size of the field of view. TVI, AC, and S-band can be estimated from videos captured by a single camera and do not require the detection of image information in detail. The proposed method is computationally efficient and simple to implement. Our motivation stems from the limited number of provided screens and a large number of monitoring cameras require the videos or image sequences ordered by change level. We demonstrate the TVI method by applying it on two video sequences, and the results show that the method can distinguish the different change levels for sequences with and without abnormal motion change; the S-band method is applied on another two video sequences and it successfully shows different sizes of the field of view.