Image-object extraction is one of the most important parts in the image processing. Object extraction is the technique of extracting objects from the pre-processed image in such a way that within – class similarity is maximized and between – class similarity is minimized. In this paper, a new method of extracting objects from grey scale static images using Fast Discrete Curvelet Transform (FDCT) via wrapping function is proposed. The motivation of using the curvelet transform in the proposed method is due to the approximate properties and the high directional sensitivity of this transform. An imaginary component of the curvelet coefficients to extract the object in the image is used. Firstly, the Curvelet transform is applied on the input image. Secondly, the Canny edge detector is applied on the edge image in all sub bands in the curvelet domain. Thirdly, the inverse of Curvelet transform is applied and finally; morphological filters are used to extract objects from the obtained binary image. Experimental results of the proposed method are compared with the results of extracting objects in the wavelet domain and the pixel domain. Indeed, the curvelet have useful geometric features that set them apart from the wavelet and the pixel domain.

Despite their high stability and compactness, chord-length features have received little attention in activity recognition literature. In this paper, we present an SVM approach for activity recognition, based on chord-length shape features. The main contribution of the paper is two-fold. We first show how a compact computationally-efficient shape descriptor is constructed using 1-D chord-length functions. Secondly, we unfold how to use fuzzy membership functions to partition action snippets into a number of temporal states. When tested on KTH benchmark dataset, the approach achieves promising results that compare very favorably with those reported in the literature, while maintaining real-time performance.