Improved grayscale projection image stabilization method suitable for tunnel environmental characteristic

Abstract

The invention relates to the technical field of traffic image processing and discloses an improved grayscale projection image stabilization method suitable for tunnel environmental characteristic. Themethod comprises the following steps of (1) acquiring a video image and establishing a reference frame; (2) performing multi-scale gaussian fitting on the video image; (3) dividing the fit video image into multiple sub-region windows; (4) performing gray scale projection of LPQ (Local Phase Quantization) characteristic-weighting on each sub-region window to obtain multiple row projection curves and column projection curves; and (5) estimating a jittering vector according to the row projection curve and the column projection curve obtained in step (4), and correcting a current image frame by use of the jittering vector. According to the improved grayscale projection image stabilization method suitable for the tunnel environmental characteristic, beginning from an actual environment of an expressway tunnel, a gray scale difference value between blurred image pixels is increased by use of the LPQ characteristic weighting with fuzzy robustness, the current frame is corrected through multi-scale Gaussian estimation and a local difference between current frame and background frame projection curves, the accuracy for overall jittering vector estimation is improved based on the concept ofmultiple sub-region windows and finally the improved grayscale projection image stabilization method suitable for the tunnel environmental characteristic is formed.

Description

technical field [0001] The invention relates to the technical field of traffic image processing, in particular to an image stabilization processing method for expressway tunnel video. Background technique [0002] Accurate extraction of traffic targets in tunnels is the key to abnormal event detection in tunnels. However, in the tunnel scene, there is vibration in the monitoring equipment, which makes the extracted vehicle target deformed, and the pedestrian target is connected to the background, which seriously interferes with the effective extraction of traffic targets. At the same time, the image quality of the tunnel image is blurred, vehicle lights and other interferences also increase the difficulty of tunnel shake video stabilization, resulting in the general effect of traditional image stabilization methods and not strong pertinence. Therefore, it is of great theoretical and practical significance to study the jitter video stabilization method for the tunnel environ...

AI Technical Summary

Problems solved by technology

Among many search strategies, the full search method has the highest search matching accuracy and can be used for comparison with other image stabilization algorithms. However, the full search method needs to calculate all search blocks in the area to be matched, which has high computational complexity and poor real-time performance, making it difficult to apply For tunnel video surveillance systems with high real-time requirements; for image stabilization methods based on bit-plane matching, such as "A Fast Bit-Plane Electronic Image Stabilization Method Based on FPGA Platform" (CN: 103179399B) applied by Harbin University, this type of method Using the binary XOR operation to speed up the calculation of the mean square error or absolute error in the search process of the block matching method, however, the selection of the bit plane is the key to this method, and different numbers and levels of bit planes will directly affect the accuracy of the jitter vector estimation and computational complexity; for the image stabilization method based on feature point matching, such as "A Electronic Image Stabilization Method for Video Cameras Based on Robust Feature Points" (CN: 107343145A) applied by Shanghai Institute of Technical Physics, Chinese Academy of Sciences, this class The method is based on the binary description operator and the matching is realized by Hamming distance calculation, which has a fast matching speed, but in the tunnel environment with light interference and blurred image quality, the false matching rate is high and the effect is not good; for grayscale projection based Image stabilization methods, such as "a grayscale projection image stabilization method and system" (CN: 107911615A) applied by Beijing Institute of Technology, this type of method replaces the two-dimensional space search in block matching and bit matching with a grayscale projection-based Two-dimensional space search has the characteristics of small amount of calculation and strong real-time performance. However, for the case where the gray gradient is relatively flat and the gray projection curve is relatively smooth, it is easy to cause multiple similar adjacent valleys in the cross-correlation graph. At the same time, when there is local illumination When there is interference, it will also cause an estimation error of the jitter vector, which cannot be well adapted to the tunnel environment

Images