Search results

The purpose of this paper is to present a new method for solving parametric programming problems; a new scheme of constraints fuzzification. In the proposed approach, constraints are learned based on deductive learning.

Adaptive neural‐fuzzy inference system (ANFIS) is used for constraint learning by generating input and output membership functions and suitable fuzzy rules.

The experimental results show the ability of the proposed approach to model the set of constraints and solve parametric programming. Some notes in the proposed method are clustering of similar constraints, constraints generalization and converting crisp set of constraints to a trained system with fuzzy output. Finally, this idea for modeling of constraint in the support vector machine (SVM) classifier is used and shows that this approach can obtain a soft margin in the SVM.

Properties of the new scheme such as global view of constraints, constraints generalization, clustering of similar constraints, creation of real fuzzy constraints, study of constraint strength and increasing the degree of importance to constraints are different aspects of the proposed method.

The purpose of this paper is to propose a lossy/lossless binary textual image compression method based on an improved pattern matching (PM) technique.

In the Farsi/Arabic script, contrary to the printed Latin script, letters usually attach together and produce various patterns. Hence, some patterns are fully or partially subsets of some others. Two new ideas are proposed here. First, the number of library prototypes is reduced by detecting and then removing the fully or partially similar prototypes. Second, a new effective pattern encoding scheme is proposed for all types of patterns including text and graphics. The new encoding scheme has two operation modes of chain coding and soft PM, depending on the ratio of the pattern area to its chain code effective length. In order to encode the number sequences, the authors have modified the multi‐symbol QM‐coder. The proposed method has three levels for the lossy compression. Each level, in its turn, further increases the compression ratio. The first level includes applying some processing in the chain code domain such as omission of small patterns and holes, omission of inner holes of characters, and smoothing the boundaries of the patterns. The second level includes the selective pixel reversal technique, and the third level includes using the proposed method of prioritizing the residual patterns for encoding, with respect to their degree of compactness.

Experimental results show that the compression performance of the proposed method is considerably better than that of the best existing binary textual image compression methods as high as 1.6‐3 times in the lossy case and 1.3‐2.4 times in the lossless case at 300 dpi. The maximum compression ratios are achieved for Farsi and Arabic textual images.

Only the binary printed typeset textual images are considered.

The proposed method has a high‐compression ratio for archiving and storage applications.

To the authors' best knowledge, the existing textual image compression methods or standards have not so far exploited the property of full or partial similarity of prototypes for increasing the compression ratio for any scripts. Also, the idea of combining the boundary description methods with the run‐length and arithmetic coding techniques has not so far been used.

The purpose of this paper is to present research in the area of the signal processing and application into pedestrian tracking in the video scene.

The paper describes the design of a new extended Kalman filter (EKF) in the high‐dimensional space (HDS) and studies of mean square error and variance analysis of error. A design algorithm is implemented in MATLAB software and tested. The data set includes many hours of captured films.

This paper includes a new derivation of the EKF and its implementation into the video scene.

The proposed algorithm can be used to track each video application.

The Kalman filter in the HDS is presented for the first time. Also, the application of the proposed method is applied in pedestrian tracking and counting.