Video Content Analysis Techniques Research Articles

Automatic defect detection for ultrasonic wave propagation imaging method using spatio-temporal convolution neural networks

Ultrasonic wave propagation imaging enables the detection of anomalies in various structures; hence, it has been applied as one of the promising techniques for damage identification in structural health monitoring (SHM). The interpretation of imaging data is vital to SHM; however, it relies significantly on expert subjective judgment, rendering the results vulnerable to human errors. Recent advances in the field of computer vision arising from the adoption of deep neural networks have resulted in new perspectives for substituting human roles in laborious data interpretation tasks. This paper presents an effective learning architecture that can characterize the ultrasonic wave propagation videos for automatic non-destructive inspection. The main contribution is threefold: 1. To the best of our knowledge, this is the first study to leverage video content analysis techniques to exploit ultrasonic wave propagation image series. Previous approaches that focused on the still wavefield images are likely to lose critical temporal information, thereby resulting in an inferior performance. 2. We devise a model that progressively aggregates both temporal and spatial information encoded in multiple adjacent snapshots of ultrasonic wave propagation motions for efficient data analysis. We presented the details regarding the system implementation and critical parameter settings. 3. The proposed approach is validated through extensive experimental comparisons with other state-of-the-art computer vision techniques on a real dataset which is publicly available. We hope that this study will encourage further investigations into video-based non-destructive data interpretation, not limited to ultrasonic signals.

VideoToVecs: a new video representation based on deep learning techniques for video classification and clustering

With the recent revolution in the field of multimedia technology, video data have become much easier and straightforward to be created, stored and transferred on a huge scale with small costs. The big amount of created data pushed the research community to delve into various study areas to aid the huge proliferation of multimedia content such as video structuring, video classification and clustering, events and objects detection, video recommendation and many other video content analysis techniques. The key success of any analysis technique relies on the audiovisual features extracted from the video. Motivated by the appearance and efficiency of deep learning techniques, we propose in this paper a new deep-learning-based features representation of videos. We depend on image-based features extracted from the sequence of frames in the video using deep learning techniques. A mapping approach named VideoToVecs is then applied to transform the extracted features into a matrix in which each row contains features of the same type. This matrix is named deep features video matrix. The efficiency of the representation is tested on 5261-video dataset for classification and clustering, and the obtained results were very promising as we will see in the paper.

Shot Classification of Field Sports Videos Using AlexNet Convolutional Neural Network

Broadcasters produce enormous numbers of sport videos in cyberspace due to massive viewership and commercial benefits. Manual processing of such content for selecting the important game segments is a laborious activity; therefore, automatic video content analysis techniques are required to effectively handle the huge sports video repositories. The sports video content analysis techniques consider the shot classification as a fundamental step to enhance the probability of achieving better accuracy for various important tasks, i.e., video summarization, key-events selection, and to suppress the misclassification rates. Therefore, in this research work, we propose an effective shot classification method based on AlexNet Convolutional Neural Networks (AlexNet CNN) for field sports videos. The proposed method has an eight-layered network that consists of five convolutional layers and three fully connected layers to classify the shots into long, medium, close-up, and out-of-the-field shots. Through the response normalization and the dropout layers on the feature maps we boosted the overall training and validation performance evaluated over a diverse dataset of cricket and soccer videos. In comparison to Support Vector Machine (SVM), Extreme Learning Machine (ELM), K-Nearest Neighbors (KNN), and standard Convolution Neural Network (CNN), our model achieves the maximum accuracy of 94.07%. Performance comparison against baseline state-of-the-art shot classification approaches are also conducted to prove the superiority of the proposed approach.

A Survey of Content-Aware Video Analysis for Sports

Sports data analysis is becoming increasingly large-scale, diversified, and shared, but difficulty persists in rapidly accessing the most crucial information. Previous surveys have focused on the methodologies of sports video analysis from the spatiotemporal viewpoint instead of a content-based viewpoint, and few of these studies have considered semantics. This study develops a deeper interpretation of content-aware sports video analysis by examining the insight offered by research into the structure of content under different scenarios. On the basis of this insight, we provide an overview of the themes particularly relevant to the research on content-aware systems for broadcast sports. Specifically, we focus on the video content analysis techniques applied in sportscasts over the past decade from the perspectives of fundamentals and general review, a content hierarchical model, and trends and challenges. Content-aware analysis methods are discussed with respect to object-, event-, and context-oriented groups. In each group, the gap between sensation and content excitement must be bridged using proper strategies. In this regard, a content-aware approach is required to determine user demands. Finally, the paper summarizes the future trends and challenges for sports video analysis. We believe that our findings can advance the field of research on content-aware video analysis for broadcast sports.

Entropy Supported Video Indexing for Content based Video Retrieval

The increased in availability and usage of on-line digital video has created a need of automated video content analysis techniques, including indexing and retrieving. Automation of indexing significantly reduces the processing cost while by minimizing tedious work. Traditional video retrieval methods based on video metadata, fail to meet technical challenges due to large and rapid growth of multimedia data, demanding effective retrieval systems. One of the most popular solutions for indexing is extracting the features of video key frames for developing a Content Based Video Retrieval (CBVR) system. CBVR works more effectively as these deals with content of video rather than video metadata. Various features like color, texture, shape can be integrated and used for video indexing and retrieval. Implemented CBVR system is experimented based on integration of texture, color and edge features for video retrieval. Entropy is a texture descriptor used for key frame extraction and video indexing. However entropy, color (RGB) and edge detection algorithms are used for video retrieval. These features are combined in various ways like entropyedge, entropycolor for result refinement. Dataset is created with the videos from different domains like e-learning, nature, construction etc. By the combination of these features in different ways, we achieved comparative results. Obtained result shows that combining of two or many features gives better retrieval.

Shot Boundary Detection for a Video under the Influence of Illumination: A Adaptive Thresholding Approach

Video Shot boundary detection is the process of automatically detecting the boundaries between shots in video. Shot Boundary detection is an important fundamental process in video data access, indexing, search and retrieval. The increased availability and usage of on-line digital video has created a need for automated video content analysis techniques. Detection of gradual transition and elimination of disturbances caused by illumination change is a major challenge in shot boundary detection technique. These disturbances are often mistaken as shot boundaries. It is a crucial task to develop a method that is not only insensitive to illumination change but also sensitive to detect shot change. An algorithm is proposed for shot boundary detection in presence of illumination change. This is very important for accurate detection of shot boundaries and very useful in content based analysis of video. First the algorithm removes illumination change using discrete cosine transform and discrete wavelet transform. Then shot boundaries are detected using normal difference & wavelet difference. A shot boundary is detected when the feature difference shows sharp change greater than threshold. Experimental study is performed on number of videos that include significant illumination change. The performance of proposed algorithm is better as compared to existing techniques

A remote markerless human gait tracking for e-healthcare based on content-aware wireless multimedia communications

Remote human motion tracking and gait analysis over wireless networks can be used for various e-healthcare systems for fast medical prognosis and diagnosis. However, most existing gait tracking systems rely on expensive equipment and take lengthy processes to collect gait data in a dedicated biomechanical environment, limiting their accessibility to small clinics located in remote areas. In this work we propose a new accurate and cost-effective e-healthcare system for fast human gait tracking over wireless networks, where gait data can be collected by using advanced video content analysis techniques with low-cost cameras in a general clinic environment. Furthermore, based on video content analysis, the extracted human motion region is coded, transmitted, and protected in video encoding with a higher priority against the insignificant background area to cope with limited communication bandwidth. In this way the encoder behavior and the modulation and coding scheme are jointly optimized in a holistic way to achieve the best user-perceived video quality over wireless networks. Experimental results using H.264/AVC demonstrate the validity and efficacy of the proposed system.