Haralick features for GEI-based human gait recognition

This paper proposes a supervised feature extraction method which is able to select discriminative features for human gait recognition under the variations of clothing and carrying conditions and hence to improve the recognition performances. The proposed method is based on the use of Haralick's texture features extracted locally from three regions of Gait Energy Images. The performance has been evaluated using CASIA Gait database (dataset B). The experimental using one-against-all SVM classifier yields attractive results when compared to existing and similar techniques.