Search

Moving object detection is a first step towards many computer vision applications, such as human interaction and tracking, video surveillance, and traffic monitoring systems. Accurate estimation of the target object’s size and shape is often required before higher-level tasks (e.g., object tracking or recog nition) can be performed. However, these properties can be derived only when the foreground object is detected precisely. Background subtraction is a common technique to extract foreground objects from image sequences. The purpose of background subtraction is to detect changes in pixel values within a given frame. The main problem with background subtraction and other related object detection techniques is that cast shadows tend to be misclassified as either parts of the foreground objects (if objects and their cast shadows are bonded together) or independent foreground objects (if objects and shadows are separated). The reason for this phenomenon is the presence of similar characteristics between the target object and its cast shadow, i.e., shadows have similar motion, attitude, and intensity changes as the moving objects that cast them. Detecting shadows of moving objects is challenging because of problem atic situations related to shadows, for example, chromatic shadows, shadow color blending, foreground-background camouflage, nontextured surfaces and dark surfaces. Various methods for shadow detection have been proposed in the liter ature to address these problems. Many of these methods use general-purpose image feature descriptors to detect shadows. These feature descriptors may be effective in distinguishing shadow points from the foreground object in a specific problematic situation; however, such methods often fail to distinguish shadow points from the foreground object in other situations. In addition, many of these moving shadow detection methods require prior knowledge of the scene condi tions and/or impose strong assumptions, which make them excessively restrictive in practice. The aim of this research is to develop an efficient method capable of addressing possible environmental problems associated with shadow detection while simultaneously improving the overall accuracy and detection stability. In this research study, possible problematic situations for dynamic shad ows are addressed and discussed in detail. On the basis of the analysis, a ro bust method, including change detection and shadow detection, is proposed to address these environmental problems. A new set of two local feature descrip tors, namely, binary patterns of local color constancy (BPLCC) and light-based gradient orientation (LGO), is introduced to address the identified problematic situations by incorporating intensity, color, texture, and gradient information. The feature vectors are concatenated in a column-by-column manner to con struct one dictionary for the objects and another dictionary for the shadows. A new sparse representation framework is then applied to find the nearest neighbor of the test image segment by computing a weighted linear combination of the reference dictionary. Image segment classification is then performed based on the similarity between the test image and the sparse representations of the two classes. The performance of the proposed framework on common shadow detec tion datasets is evaluated, and the method shows improved performance com pared with state-of-the-art methods in terms of the shadow detection rate, dis crimination rate, accuracy, and stability. By achieving these significant improve ments, the proposed method demonstrates its ability to handle various problems associated with image processing and accomplishes the aim of this thesis.

An accurate estimation of the target object size and shape is an essential step in many computer vision applications, including object tracking and object recognition. However, due to the presence of cast shadow, these properties cannot be obtained accurately using common object detection systems. In many cases, due to having similar characteristics, shadow points are misclassified as parts of moving objects. To deal with the problem of misclassifying shadows as foreground, various methods have been introduced. Most of these methods have been designed to detect shadows in specific situations and often fail to distinguish shadow points from the foreground object in many cases, including foreground-background camouflage and foreground-shadow camouflage. The aim of this research is to solve these two environmental problems and to improve accuracy and real-time performances by developing two effective algorithms for detecting moving shadows in image sequences. The first algorithm uses three features, namely, intensity, invariant colour constancy and temporal colour constancy, to distinguish shaded regions from objects. In this method, an initial clustering of the change detection mask is used to divide the mask into sub-regions, then three quantities, namely, intensity mean, invariant colour constancy measurement and temporal color constancy measurement are computed for each region. Initial classification is made based on these three measurements followed by inter-region relationships among neighbouring regions to enhance the final detection result. In this method, a number of improvements in the shadow detection systems have been introduced. First, a new model of colour constancy, namely, invariant colour constancy, has been analysed and formulated. Invariant colour constancy model has been shown its efficiency in presence of camouflages. Second, temporal colour constancy has been modelled to sustain high levels of shadow detection accuracy. In particular, temporal colour constancy works well in presence of foreground-shadow camouflage. Finally, inter-region dependencies among neighbouring sub-regions are used to eliminate regions with false positives and negatives. The second method is designed to detect moving shadows of vehicles in real-time applications. The method is based on two measurements, namely, the illumination direction and the intensity measurements in the neighbouring pixels in a scanned line, to detect intervals with a decreasing function. This proposed method introduces three innovative concepts. Firstly the proposed method works on image-line analysis, which is different from traditional methods relying on either pixel-based or region-based analysis. Secondly it exploits two features, namely, spatial and intensity relationships, among points belonging to the scanned image-line to detect intervals with a decreasing function. This is done by using the illumination direction of the dominant light source and the amount of the receiving energy of the sunlight by each point. Finally, the linear property of the transition factor inside the shadow cast area is proposed as a new shadow detection criterion. Experimental results show better performances of the two proposed methods over other methods and better achievements in terms of detection rate, shadow discrimination rate and efficiency in real-time performances when compared to existing methods. These significant improvements show that the methods developed in this research meet the aim of this thesis.