Video Object Tracking Method Based on Snake Model Using Object´s Histogram Information

This paper presents a video object tracking algorithm based on snake model using object´s histogram information, which is designed on the basis of object´s histogram consistency between the consecutive image frames, and finally implemented by using the level set method. Because this algorithm uses the histogram information in the snake energy function, it overcomes the local minimization problem of video tracking algorithms based on the local characters, such as mean, deviation or motion vector. The contrastive experiment reveals that our algorithm performs better than those based on local characters