杂波背景下慢动目标多特征融合检测系统设计

Design of multi feature fusion detection system for slow moving target in clutter background

摘要:

杂波背景中的干扰因素众多，导致慢动目标检测精度低、效率差，提出了一种多特征融合检测系统设计方案。硬件部分由预处理模块、模数转换模块和DSP模块三部分组成，利用有源低通滤波抑制了高频噪声的干扰，通过DSP实现对其他模块的有效控制；软件部分由多特征融合检测模块、通信接口模块以及存储模块构成，将包含慢动目标的原始图像划分为若干个子区域，并确定每个子区域的CLBP直方图和色调直方图，与背景模型进行相似性匹配，实现慢动目标多特征融合检测；通过USB接口将检测结果传送至存储模块中，完成检测结果的存储。系统性能测试实验结果表明，设计系统可在花费最少时间的前提下，得到精度最高的杂波背景下慢动目标检测结果。

Abstract:

There are many interference factors in clutter background, which leads to low accuracy and poor efficiency of slow moving target detection. A design scheme of multi feature fusion detection system is proposed. The hardware part is composed of preprocessing module, analog-to-digital conversion module and DSP module. Active low-pass filter is used to suppress the interference of high-frequency noise, and DSP is used to realize the effective control of other modules; The software part is composed of multi feature fusion detection module, communication interface module and storage module. The original image containing slow moving target is divided into several sub regions, and the CLBP histogram and tone histogram of each sub region are determined to match with the background model to realize the multi feature fusion detection of slow moving target; The test results are transmitted to the storage module through the USB interface to complete the storage of the test results. The experimental results of system performance test show that the designed system can obtain the slow-moving target detection results with the highest accuracy under the clutter background with the least time.