Majlesi Journal of Multimedia Processing
Volume:2 Issue: 3, Sep 2013

High Efficiency Video Coding (HEVC) [3] is the next generation video compression standard being jointly developed by the Joint Collaborative Team on Video Coding (JCT-VC) of ITU-T WP3/16 and ISO/IEC JTC 1/SC 29/WG 11. The loss less coding is prominent in real-time applications like automotive vision, video conferencing and in web collaboration for remote desktop sharing. HEVC with the lossless mode can help in these applications effectively by providing content with certain level of compression when compared to other lossless compression techniques. This paper focuses on improving the compression efficiency for the lossless coding mode of HEVC by using the approach of sample based angular intra prediction [1], replacing the traditional intra prediction approach currently used in HEVC. The sample based angular intra prediction approach uses the same prediction mode signaling method and the same interpolation method as the HEVC block based angular intra prediction but instead uses adjacent neighbors as the reference samples for better intra prediction accuracy and performs prediction pixel by pixel. Compared to the HEVC-anchor mode (HM 9.2 [4]) the proposed SAP [1] based lossless mode in this paper achieves significant bit rate savings from 5.93% - 12.76% for AI configuration and 2.56% - 7.87% for LB-Main configuration. It also increases compression ratio by 10.7% for AI and 5.3% for LB-Main configurations respectively. The encoding and decoding times are also reduced using the SAP based HEVC lossless mode [1]. For implementation purpose HM 9.2 [4] version of the HEVC reference software is used and the current HEVC draft is followed to comply with the semantics of the software. The proposed method is compared with existing lossless compression techniques such as JPEG-2000 [8], JPEG-LS [7], 7-Zip [9] and Win RAR [10].

In this paper there are guidelines to decrease fault alerts in motion detection algorithms in videos. we prevent from electrical noise detection and the leaves motion and unnecessary motions instead of the subject in pictures. For this purpose we must first detect kinds of motions in picture then by proposed algorithms from concern subjects.

Lane detection is one important subjects of recent researches which is done by using different algorithms. the used system in this algorithm have been acquired (or taken) from the human brain processing system which simoultaneously uses both high-speed short-sighted processing for deciding about near vision and low-speed far-sighted processing for estimating the distance and future decisions.The lower section of the picture (lower cut half) is used for lane detection and the placement in relation wite them using a series of processing based on lands and high speeds. Also,the upper section of the picture (upper cut half) in this algorithm is used to estimate the distance and vanishing point recognition of the road that provides the system with further necessary information about the continuation.These two systems work in parallel with each other but with different processing speeds to analyze the information. In the first part, which is lane detection and tracking it after some pre-processing operations including picture quality improvement, edge detection, and cutting the upper half of the picture to delete extra objects, using some characteristics related to road lanes such as lane directions, length to width relatione, color, occupied area and lane angles extra parts are removed and main road lanes are recognized with houghtransformation.Estimation and vanishing point recognition, a simple algorithm is given to find and follow road vanishing points avoiding any In the second part, which is related to road complicated calculations to use in current processors.This method is generally based on the estimation of vanishing points places in relation to a vehicle. The algorithm which is used to find these points is very efective for noise peaks,infinite place and irrelevant lines.

In this paper, we propose a fast image classifier motivated from emotional process in the brain. Our model is named Brain Emotional Learning based Picture Classifier (BELPIC) and is modified version of Brain Emotional Learning (BEL) applied in various control applications. The distinctive feature of BELPIC is applying the activation function hardlim in the model. Here, various comparisons are made between BELPIC (available from www.bitools.ir) and multilayer perceptron (MLP) to classify the images of a typical dataset. According to the experimental results, BELPIC shows higher accuracy and lower time complexity in image classification.

Digitization in every sector leads to the growth of digital data in a tremendous amount. Digital data are not only available in the form of text but it is also available in the form of images, audio and video. Decision making people in every field like business, public sector, hospital, etc. are trying to get useful and implicit information from the already existing digital data bases. Image mining is the concept used to extract implicit and useful data from images stored in the large data bases. Image mining is used in variety of fields like medical diagnosis, space research, remote sensing, agriculture, industries and even in the educational field. This paper elaborates the research works already done in image mining and also summarizes different tool developed, algorithms emerged and the applications of image mining used to extract the useful images in various fields.

This paper proposed a new biometric system based on the texture of hand knuckles. The texture pattern produced by finger knuckle bending is highly unique and makes the surface a distinctive biometric identifier. In this approach, we used Contourlet and wavelet transforms to extract features from finger knuckle surfaces. Then we used AGV to select the best features. Finally we examined three famous classifier to select the one with the best results. The experimental results confirmed the 97.66% recognition rate for the proposed system.

A Data Grid consists of a collection of geographically distributed computer and storage resources located in different places, and enables users to share data and other resources. Data replication is used in Data Grid to enhance data availability, fault tolerance, load balancing and reliability. Although replication is a key technique, but the problem of selecting proper locations for placing replicas i.e. replica placement in Data Grid has not been widely studied, yet. In this paper an efficient replica selection strategy is proposed to select the best replica location from among the many replicas. Due to the limited storage capacity, a good replica replacement algorithm is needed. We present a novel replacement strategy which deletes files in two steps when free space is not enough for the new replica. A Grid simulator, OptorSim, is used to evaluate the performance of this dynamic replication strategy. The simulation results show that the proposed algorithm outperforms comparing to than the existing algorithms mentioned in literature as number of jobs increases.