OpenCV Based On Gait Recognition System

Human activity analysis has attracted great interest from computer vision researchers due to its promising applications in many areas such as automated visual surveillance, computer-human interactions, and motion-based identification and diagnosis.Gait features are unique. Gait recognition is one of General biological characteristics, it features includes long-range, non-contact, difficult hide and so on. It overcomes the traditional pattern recognition (iris recognition, fingerprint recognition, etc.) in high-resolution and close range. Gait recognition has a good recognition in the case of low resolution results, therefore, it has a wide applied future in intelligent video surveillance. This article mainly studies gait video images provided by the CAS Institute of Automation, using OpenCV to build a gait recognition system. The following are its main tasks.First, the extraction of human object. Because of the complex background, we establish a stable background model by using the adaptive background modeling and get the goal of human motion by using background subtraction. As for the existence of the shadow of the human body and inanity, this article makes shadow detection and elimination by using color space conversion respectively and handles human target image soothingly by using regional filling and morphological filtering on smoothing.Second, the gait feature extraction. Because the number of high-dimensional video images is high, this article adopts a gait energy image, which is a way of turning gait cycle into a frame image. Gait energy image not only saves storage space and computing time, but also is not sensitive to the noise on the single frame. It uses the (2D)2PCA features extracted to reduce the dimensions so as to solve the curse of dimensionality, and then analyzes the contribution of value of Gait features given by the gait characteristics of each feature vector with weighted (2D)2 PCA. Thus it well defined the main features of vector affecting gait recognition rate.Third, the gait identification. It analyzes the HMM algorithm and makes use of HMM to classification training of Gait features extracted, then the classification results are analyzed.Fourth, the design and implementation of system. This gait recognition system is achieved loading OpenCV under VC++6.0 visual library. Experimental samples adopt different perspective video library provided by the CAS Institute of Automation of data, and gait recognition algorithm used experimentally in the system is performed on experiment. The system design has clear features and easy operation.