基于改进CoHOG-LQC的行人检测算法

Extended CoHOG-LQC for Pedestrian Detection

针对行人检测过程中,易对相似目标产生误判的问题,并结合局部纹理特征描述子对图像边缘、方向信息的描述能力与检测精度的强相关性,同时考虑到基于LBP和HOG的特征融合方法存在结构利用率低、光谱信息损失多的缺点,提出了一种基于LQC和CoHOG特征融合的行人检测算法。首先通过LQC算子提取图像的纹理谱特征,同时使用积分图计算CoHOG特征值,以提取原始图像的边缘特征及基于LQC特征谱的CoHOG特征。然后将上述特征与CoHOG边缘特征融合,得到融合特征描述图像,最后使用HIKSVM分类器实现输入图像的检测与识别。为验证算法的有效性,分别在MIT行人数据库、Caltech行人数据库和INRIA行人数据库上进行实验。实验结果表明,提出的方法可以有效提高行人检测精度和效率。

In accordance with the miscalculation over the recognition of resemble objects in the process of pedestrian detection, and strong correlations between detection precision and description capability that local texture feature descriptors can achieve when acquiring the characteristics of image edge and direction, considering the de- fects that the low space efficiency as well as high spectral information loss of the pedestrian tracking algorithm which based on fusion among Local Binary Pattern （LBP） and Histograms of Oriented Gradient （HOG）. We proposed a novel algorithm based on the fusion among Local Quantization Code （LQC） feature and Co-Occurrence Histogram Oriented Gradient （CoHOG） feature for detecting passenger. Firstly, the spectral property of the image were extrac- ted efficiently using LQC feature descriptor from image. Next, the calculation using integral image was established to withdraw edge characteristic and CoHOG features based on LQC character spectrums from the original image. For further procedure, the CoHOG edge feature are fused with them, then the fusion feature image is acquired. At last, Histogram Intersection Kernel Support Vector Machine （HIKSVM） classifiers were performed for detection and rec- ognition. To validate the effectiveness of the algorithm, experiments are carried out on 3 public pedestrian dataset including MIT, Caltech and INRIA. The results demonstrates that the method is effective to raise accuracy and effi- ciency of clustering process.