No abstract

No abstract

The recognition of handwritten characters and digits is an important and challenging issue in OCR algorithms. This article presents a new method in which cluster based weighted support vector machine is used for the classification and recognition of Farsi handwritten digits that is reasonably robust against rotation and scaling. In the proposed algorithm, after applying the necessary preprocessing on the digits images, the required features are extracted using principle component analysis (PCA) and linear discrimination analysis (LDA) algorithms. The extracted features are then classified using a new classification algorithm called cluster based weighted SVM (CBWSVM). We tested the proposed algorithm with a database containing 7600 handwritten digits with and without rotation and the results showed the recognition rate of 96.5% in digits without rotation and 95.6% in digits with rotation of the 15 degrees. The comparison of the results with those of other methods showed the efficiency of the proposed algorithm. Index Terms-CBWSVM, Clustering, Handwritten digit recognition, PCA, PCA-LDA .Alireza Behrad is with the Faculty

In this article, a new method for the recognition of obscene video contents is presented. In the proposed algorithm, different episodes of a video file starting by key frames are classified independently by using the proposed features. We present three novel sets of features for the classification of video episodes, including (1) features based on the information of single video frames, (2) features based on 3D spatiotemporal volume (STV), and (3) features based on motion and periodicity characteristics. Furthermore, we propose the connected components' relation tree to find the spatiotemporal relationship between the connected components in consecutive frames for suitable features extraction. To divide an input video into video episodes, a new key frame extraction algorithm is utilized, which combines color histogram of the frames with the entropy of motion vectors. We compare the results of the proposed algorithm with those of other methods. The results reveal that the proposed algorithm increases the recognition rate by more than 9.34% in comparison with existing methods.