A semantic video analysis system is a semi- or fully-automated system for investigation of video contents and extraction of concepts and semantics from video. These systems can be classified from different viewpoints. In this article, the semantic gap between low-level features and high-level concepts in the semantic video analysis systems is discussed and the literature is reviewed from viewpoint of semantic hierarchy for video production.To this end, after a brief description of video analysis systems and their general block diagram, two main challenges of these systems are discussed: sensory gap and semantic gap. Then, different approaches for reduction of the semantic gap are studied based on a semantic hierarchy used for video production. According to this hierarchical structure, there are three main steps for reduction of semantic gaps: frame processing; content analysis; and semantic extraction. Finally, open problems in this field of research are presented as the following. Usually, a wide variety of events and concepts may occur in a video. Different concepts may be assigned to an event in different circumstances. Some high-level concepts may occur in a relatively long duration of video; thus, extraction of such concepts requires processing of a relatively long duration of video to construct a semantic network between the concepts extracted from short durations of video. Usually, analysis of multi-modal information may close the semantic gaps that exist in the analysis of single-modal information.

The aim of this study was the video analysis of soccer player's injuries in the 2010 World Cup games. Therefore, all World Cup games were reviewed through video analysis and their information (the number of injuries, the mechanism of injuries, the time of incidence, the type of accident leading to injuries and the injury occurrence area in the pitch) were recorded in the injury report form. In the present study that was descriptive-analytical, Chi square test (X2) was used for statistical analysis. Results showed that 92 injuries occurred during 64 games or 39.3 injuries per 1000 hours of game. Most incidence of injuries occurred in the final 15 minutes of the second half (minutes of 76-90) (P

Analyzing motion patterns in traffic videos can directly lead to generate some high-level descriptions of the video content. In this paper, an unsupervised method is proposed to automatically discover motion patterns occurring in traffic video scenes. For this purpose, based on optical flow features extracted from video clips, an improved Group Sparse Topical Coding (GSTC) framework is applied for learning semantic motion patterns. Then, each video clip can be sparsely represented by a weighted sum of learned patterns which can further be employed in very large range of applications. Compared to the original GSTC, the proposed improved version of GSTC selects only a small number of relevant words for each topic and hence provides a more compact representation of topic-word relationships. Moreover, in order to deal with large-scale video analysis problems, we present an online algorithm for improved GSTC which can not only deal with large video corpora but also dynamic video streams. Experimental results show that our proposed approach finds the motion patterns accurately and gives a meaningful representation for the video.

Analyzing motion patterns in traffic videos can be employed directly to generate high-level descriptions of their content. For traffic videos captured from intersections, usually, we can easily provide additional information about traffic phases. Such information can be obtained directly from the traffic lights or through traffic lights controllers. In this paper, we focus on incorporating additional information to analyze the traffic videos more efficiently. Using side information on traffic phases, the semantic of motion patterns from traffic intersection scenes can be learned more effectively. The learning is performed based on optical flow features extracted from training video clips, and applying them to supervised topic models such as MedLDA and MedSTC. Based on such models, any video clip can be represented based on the learned patterns. Such representations can be further exploited in scene analysis, rule mining, abnormal event detection, etc. Our experiments show that employing side information in intersection video analysis leads to improvement in discovering scene pattern. Moreover, supervised topic models achieve about 4% improvement in abnormal event detection, compared to the unsupervised ones, in terms of area under ROC.

In this paper, a novel method for video content summarization has been proposed by calculating the fractal dimension of frames. Summarization of the video is the first step in automatic video analysis. In this paper, we use the support vector machine (SVM) and the decision - tree to identify the shot boundary and classify them. In order to compute the fractal dimension, the numerical method has been expressed. The results of this implementation were also reported on TRECVID 2006 data collection. The results show that the relative advantage of the method presented in this paper is compared to other articles.

The current study aimed to explore the effectiveness of a process based reflective teacher training course that utilized situational role plays, video analysis, and scenario evidence as guided reflective and caring practices for EFL (English as a Foreign Language) teacher candidates. To this end, 45 teacher candidates at a teacher education university in Iran participated in a semi-structured interview (pre-intervention and post-intervention interview) and created process-oriented narrative journals documenting their reflective and caring practices. The findings of the study indicated that all the three groups of participants (role play group, scenario-based evidence group, and video evidence group) benefitted from the course and demonstrated improvement in various ways. The emerged themes from each group (role play group: 23 categories, video group: 17 categories, and scenario-based evidence group: 16 categories) illustrated how these strategies supported teacher candidates in developing their caring and reflective practices.

Despite the acknowledged impact of imagery on performance, ecologically sound studies investigating imagery and its effects on performance subcomponents in real games are surprisingly limited. The aim of this study was to investigate the effects of imagery training on passing improvement in elite soccer players. Sixty nine soccer players taking part in the national championship leagues in four age categories including U16, U19, U21 and over 21 randomly assigned to the imagery and control groups. Interventional group participants completed 8 weeks, video-aided, cognitive imagery program on how to make a perfect soccer pass. Performance analysis through close video analysis showed that successful pass rate increased significantly in intervention group compared to control (OR=1.19, P=0.002, 95%CI=1.06, 1.33). Further analysis revealed that the results are statistically significant in U16 and U21 but not other categories. We concluded that successful soccer passing through real competitions as a multidimensional and critical open skill could be enhanced by an ecologically sound method of mental imagery.

Video text detection plays an important role in applications such as semantic-based video analysis، text information retrieval، archiving and so on. In this paper، we propose a Farsi/Arabic text detection approach. First، with an appropriate edge detector، edges are extracted and then by using edges cross ponts، artificial corners are extracted. Artificial corner histogram analysis is done for rejecting some non text corners. The discrete cosine transform (DCT) coefficients of input picture are extracted and texture intensity picture is created by combining appropriate coefficients. With combining artificial corners image and texture intensity image، a features vector is extracted and fed into support vector machine (SVM) classifier for detecting text regions. Finally with drawing normalized texture intensity profiles، final verification is done and text lines are sepersted from each others

Warning systems are developed for left-turning vehicles at intersections where protected left-turns are not warranted or cannot be provided, due to limitations of right of way or intersection capacity. These are meant to provide warnings to left-turning vehicles of the approaching vehicles from the opposite direction, when the time to turn deems unsafe. To implement such systems, real time estimation of the probability of vehicle conflict is necessary. A study has been conducted using video and radar at various intersections, to obtain estimates of turning time and acceptable gaps for drivers under various situations. Initial pilot observations indicate that the presence of pedestrians in intersections had an immediate and substantial impact on the movements of left-turning vehicles. From a preliminary systematic video analysis, in the presence of pedestrians on the destination crosswalk, the mean and standard deviation of both the gap and gap components (i.e., turning time and the “buffer” between the turning vehicle and the next oncoming vehicle) increases. On the basis of these observations, pedestrian-detection mechanisms would be useful for such intersection warning systems with a threshold for warning that is adjusted for pedestrians when they are present or in the vicinity of the destination crosswalk.

The purpose of this study was to Video analysis of sports injuries in the Iranian Football Premier League before and after the outbreak of the Corona virus in the 2019-2020 season. The video of 143 matches out of 239 matches of these games was reviewed by the researcher and the prevalence of injury, the mechanism of injury, the position of the injured player and the area of injury were recorded. According to the results, 219 injuries occurred before the outbreak of coronavirus and 276 injuries occurred after the outbreak of coronavirus. Findings showed that despite the significant increase in the amount of damage in the corona period compared to the previous period of the corona, there was no significant difference in the mechanism of injury, the area of injury and the position of the injured player, before and after the corona virus outbreak.