Chia-Hung Yeh

Moving cast shadow elimination for robust vehicle extraction based on 2D joint vehicle/shadow models

A new algorithm to eliminate moving cast shadow for robust vehicle detection and extraction in a vision-based highway monitoring system is investigated. The proposed algorithm is based on a simplified 2D vehicle/shadow model of six types projected on to a 2D image plane. Parameters of the joint 2D vehicle/shadow models can be estimated from the input video without light source and camera calibration information. Simulations are performed to verify that the proposed technique is effective for vision-based highway surveillance systems.

Robust vehicle and traffic information extraction for highway surveillance

A robust vision-based traffic monitoring system for vehicle and traffic information extraction is developed in this research. It is challenging to maintain detection robustness at all time for a highway surveillance system. There are three major problems in detecting and tracking a vehicle: (1) the moving cast shadow effect, (2) the occlusion effect, and (3) nighttime detection. For moving cast shadow elimination, a 2D joint vehicle-shadow model is employed. For occlusion detection, a multiple-camera system is used to detect occlusion so as to extract the exact location of each vehicle. For vehicle nighttime detection, a rear-view monitoring technique is proposed to maintain tracking and detection accuracy. Furthermore, we propose a method to improve the accuracy of background extraction, which usually serves as the first step in any vehicle detection processing. Experimental results are given to demonstrate that the proposed techniques are effective and efficient for vision-based highway surveillance.

Fast Mode Decision Algorithm for Scalable Video Coding Using Bayesian Theorem Detection and Markov Process

The newest video coding standard called scalable video coding (SVC) provides broad applications in multimedia communications. SVC encoder consumes great computational complexity when compared to previous video coding standards. This paper presents a fast mode decision algorithm that speeds up the SVC encoding process through probabilistic analysis. The mode of the enhancement layer is first predicted by statistical analysis. Afterward, Bayesian theorem is utilized to detect whether the prediction mode of the current macroblock is the best or not. The mode is further predicted and refined by the Markov process. Experimental results show that the proposed algorithm significantly reduces computational complexity with negligible peak signal-to-noise ratio degradation and bitrate increase in the enhancement layers.

Haze effect removal from image via haze density estimation in optical model

Images/videos captured from optical devices are usually degraded by turbid media such as haze, smoke, fog, rain and snow. Haze is the most common problem in outdoor scenes because of the atmosphere conditions. This paper proposes a novel single image-based dehazing framework to remove haze artifacts from images, where we propose two novel image priors, called the pixel-based dark channel prior and the pixel-based bright channel prior. Based on the two priors with the haze optical model, we propose to estimate atmospheric light via haze density analysis. We can then estimate transmission map, followed by refining it via the bilateral filter. As a result, high-quality haze-free images can be recovered with lower computational complexity compared with the state-of-the-art approach based on patch-based dark channel prior.

Region-of-interest video coding based on rate and distortion variations for H.263+

Region-of-interest (ROI) is an essential task that one must undertake in low bit-rate multimedia communications because of the limited bandwidth of the channels and the transcoder between different standards. In this paper, an effective ROI determination method is proposed. The skin-color extraction is first employed to determine the ROI macroblocks; then, the framework for the adjustment of the quantization parameters (QPs) of ROI macroblocks according to the distortion and bit-rate variations is proposed to fit the target bit rate. In view of the residual distortion information, the fuzzy logic controller proposed in our framework can adaptively adjust the weighting factor for corresponding QPs by the distortion variation in the macroblock layer. A linear prediction formula, derived from the rate variation, is proposed to allocate appropriate bits for each ROI macroblock and maintain the target bit rate and buffer fullness. Experiment results on TMN8 reveal that the proposed ROI video coding can significantly enhance the quality at ROIs. Furthermore, the proposed framework obtains about 0.5dB gain in the objective performance and also provides the better subjective quality compared with previous works.

Real-time background modeling based on a multi-level texture description

Background construction is the base of object detection and tracking of machine vision systems. Traditional background modeling methods often require complicated computations and are sensitive to illumination changes. This paper proposes a novel block-based background modeling method based on a hierarchical coarse-to-fine texture description, which fully utilizes the texture characteristics of each incoming frame. The proposed method is efficient and can resist both illumination changes and shadow disturbance. The experimental results show that this method is suitable for real-world scenes and real-time applications.

Fast Mode Decision Algorithm Through Inter-View Rate-Distortion Prediction for Multiview Video Coding System

Multiview video coding (MVC) has attracted great attention from industries and research institutes. MVC is used to encode stereoscopic video streams for 3D playout systems such as 3D television, digital cinema, and IP network applications. MVC is an extended version of H.264/AVC that improves the performance of multiview videos. Yet, when compared with single-view video coding, MVC consumes much more time when encoding large amounts of data. Speed-up algorithms, therefore, are essential for realizing related applications. This paper presents a fast mode decision algorithm to avoid the high computational complexity of MVC. The proposed approach aims to reduce candidate modes and make mode decision process more efficient. The minimum and maximum values of rate-distortion cost (RD cost) in the previously encoded view are used to compute a threshold for each mode in the current view. Compared with joint multiview video coding, the experimental results demonstrate that the proposed algorithm provides an average of 79% in time savings with negligible bit rate increase and peak signal-to-noise ratio decrease.

Self-learning-based post-processing for image/video deblocking via sparse representation

Abstract Blocking artifact, characterized by visually noticeable changes in pixel values along block boundaries, is a common problem in block-based image/video compression, especially at low bitrate coding. Various post-processing techniques have been proposed to reduce blocking artifacts, but they usually introduce excessive blurring or ringing effects. This paper proposes a self-learning-based post-processing framework for image/video deblocking by properly formulating deblocking as an MCA (morphological component analysis)-based image decomposition problem via sparse representation. Without the need of any prior knowledge (e.g., the positions where blocking artifacts occur, the algorithm used for compression, or the characteristics of image to be processed) about the blocking artifacts to be removed, the proposed framework can automatically learn two dictionaries for decomposing an input decoded image into its “blocking component” and “non-blocking component.” More specifically, the proposed method first decomposes a frame into the low-frequency and high-frequency parts by applying BM3D (block-matching and 3D filtering) algorithm. The high-frequency part is then decomposed into a blocking component and a non-blocking component by performing dictionary learning and sparse coding based on MCA. As a result, the blocking component can be removed from the image/video frame successfully while preserving most original visual details. Experimental results demonstrate the efficacy of the proposed algorithm.

Robust Region-of-Interest Determination Based on User Attention Model Through Visual Rhythm Analysis

Region-of-interest (ROI) determination is very important for video processing and it is desirable to find a simple method to identify the ROI. Along this direction, this paper investigates a user attention model based on visual rhythm analysis for automatic determination of ROI in a video. The visual rhythm, which is an abstraction of a video, is a thumbnail version of a video by a 2-D image that captures the temporal information of a video sequence. Four sampling lines, including diagonal, anti-diagonal, vertical, and horizontal lines, are employed to obtain four visual rhythm maps in order to analyze the location of the ROI from video data. Via the variation on visual rhythms, object and camera motions can be efficiently distinguished. As for hardware design consideration, the proposed scheme can accurately extract ROI with very low computational complexity for real-time applications. The promising results from the experiments demonstrate that the moving object is effectively and efficiently extracted. Finally, we present a way to use flexible macroblock ordering in combination with ROI determination as a preprocessing step for H.264/AVC video coding, and experimental results show the quality of ROI regions is significantly enhanced.

Learning-Based Joint Super-Resolution and Deblocking for a Highly Compressed Image

A highly compressed image is usually not only of low resolution, but also suffers from compression artifacts (blocking artifact is treated as an example in this paper). Directly performing image super-resolution (SR) to a highly compressed image would also simultaneously magnify the blocking artifacts, resulting in an unpleasing visual experience. In this paper, we propose a novel learning-based framework to achieve joint single-image SR and deblocking for a highly-compressed image. We argue that individually performing deblocking and SR (i.e., deblocking followed by SR, or SR followed by deblocking) on a highly compressed image usually cannot achieve a satisfactory visual quality. In our method, we propose to learn image sparse representations for modeling the relationship between low- and high-resolution image patches in terms of the learned dictionaries for image patches with and without blocking artifacts, respectively . As a result, image SR and deblocking can be simultaneously achieved via sparse representation and morphological component analysis (MCA)-based image decomposition. Experimental results demonstrate the efficacy of the proposed algorithm.