Study Of Battle Damage Effect Assessment Based On Image Change Detection

BDA（Battle Damage Assessment）based on Image Process is a difficulty inmilitary research, domestic scholars began to study this field is relatively late, thus thereare little theoretical framework and guidance of the system. In the process ofimplementing a military strike, damage effect evaluation can provide strong decisionsupport for commanders, the low efficiency of traditional damage assessment whichwere manually evaluated cannot meet the demand of Information War. Thus conductingDamage Assessment based on computer image processing and analysis has very vitalsignificance in improving the efficiency of evaluation, saving battle materials andstrengthening the forces fighting capacity. This is also what will be focused in thispaper.First of all, according to the analysis of a large number of literature study and thebasic damage assessment technologies based on Image Processing, this paper constructsthe evaluation model on the basic of change detection scheme. And then in turn do atheoretical analysis of Image Preprocessing, Image change detection, Data Fution andEvaluation classification system. At the same time, base on the basic theory, detailedoperation steps of each processing stage algorithm are given in thin paper, then thealgorithms are analyzed and verified through simulation. Aiming at the shortcomings ofthe traditional change detection algorithm, this paper puts forward some improvedalgorithm. First, in view of the unsounded testing effect of classical change detectionalgorithm s, this paper puts forward a combination method based on wavelet multi-scaledecomposition, difference PCA analysis and difference fusion. Second, in view of theinefficient use of the neighborhood of the change pixel and the correlation betweendifferent results, this paper puts forward a improved measurement algorithm based onmulti-scale difference PCA detection and MRF fusion. Finally, according to the changemeasurement results, this paper attempts to construct the evaluation classification model,then carry on the grading evaluation on a sample picture. In this paper, the simulationresults of each steps prove the feasibility and effectiveness of the preset detectionscheme, at the same time prove it has certain practicality.