• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.431-433
• /
• 2021

In this paper, abnormal behavior monitoring system using YOLO AI platform was implemented and had superior response characteristics compared to the conventional monitoring system using two-shot detection by using one-shot detection of YOLO system. The YOLO platform was trained using image dataset composed of abnormal behaviors such as assault, theft, and arson. The abnormal behavior monitoring system consists of client and server and can be applicable to smart cities to solve various crime problems if it is commercialized.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.1
• /
• pp.133-138
• /
• 2008

In this paper, we propose a novel anchor shot detection system, named to MASD (Multi-phase Anchor Shot Detection), which is a core step of the preprocessing process for the news video analysis. The proposed system is composed of four modules and operates sequentially: 1) skin color detection module for reducing the candidate face regions; 2) face detection module for finding the key-frames with a facial data; 3) vector representation module for the key-frame images using a non-negative matrix factorization; 4) one class SVM module for determining the anchor shots using a support vector data description. Besides the qualitative analysis, our experiments validate that the proposed system shows not only the comparable accuracy to the recently developed methods, but also more faster detection rate than those of others.

• Journal of KIISE:Software and Applications
• /
• v.36 no.2
• /
• pp.169-177
• /
• 2009

In this paper, we propose a new shot boundary detection method which is optimized for news video story parsing. This new news shot boundary detection method was designed to satisfy all the following requirements: 1) minimizing the incorrect data in data set for anchor shot detection by improving the recall ratio 2) detecting abrupt cuts and gradual transitions with one single algorithm so as to divide news video into shots with one scan of data set; 3) classifying shots into static or dynamic, therefore, reducing the search space for the subsequent stage of anchor shot detection. The proposed method, based on singular value decomposition with incremental clustering and mercer kernel, has additional desirable features. Applying singular value decomposition, the noise or trivial variations in the video sequence are removed. Therefore, the separability is improved. Mercer kernel improves the possibility of detection of shots which is not separable in input space by mapping data to high dimensional feature space. The experimental results illustrated the superiority of the proposed method with respect to recall criteria and search space reduction for anchor shot detection.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.54 no.11
• /
• pp.669-672
• /
• 2005

An algorithm detecting abrupt and gradual shot boundaries is proposed in this Paper. The conventional methods detect abrupt shot boundaries well, but do not show good performance on gradual shot boundaries. The proposed method Is based on the fact that the difference of the characteristic between frames is large when the shot conversion occurs. And the Proposed method detects abrupt and gradual shot boundaries with one algorithm. Moreover, it detects not only position where gradual shot conversion occurs, but also the exact duration where gradual shot conversion occurs.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.12 no.8
• /
• pp.3670-3676
• /
• 2011

A cell animation is represented by one background cell, and there is much difference of images when its shot is changed. Also, it does not have a lot of colors since people themselves draw it. In order to effectively detect shot transitions of cell animations while fully considering their intrinsic characteristics, in this paper, we propose a animation shot boundary detection algorithm that utilizes color and block-based histograms step by step. The suggested algorithm first converts RGB color space into HSI color one, and coarsely decides if adjacent frames contains a shot transition by performing color difference operation between two images. If they are considered to have a shot transition candidate, we calculate color histograms for 9 sub-regions of the adjacent images and apply weights to them. Finally, we determine whether there is a real shot transition by analyzing the weighted sum of histogram values. In experiments, we show that our method is superior to others.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.40 no.4
• /
• pp.96-104
• /
• 2017

A one-shot system (device) refers to a system that is stored for a long period of time and is then disposed of after a single mission because it is accompanied by a chemical reaction or physical destruction when it operates, such as shells, munitions in a defense weapon system and automobile airbags. Because these systems are primarily related with safety and life, it is required to maintain a high level of storage reliability. Storage reliability is the probability that the system will operate at a particular point in time after storage. Since the stored one-shot system can be confirmed only through inspection, periodic inspection and maintenance should be performed to maintain a high level of storage reliability. Since the one-shot system is characterized by a large loss in the event of a failure, it is necessary to determine an appropriate inspection period to maintain the storage reliability above the reliability goal. In this study, we propose an optimal inspection policy that minimizes the total cost while exceeding the reliability goal that the storage reliability is set in advance for the one-shot system in which periodic inspections are performed. We assume that the failure time is the Weibull distribution. And the cost model is presented considering the existing storage reliability model by Martinez and Kim et al. The cost components to be included in the cost model are the cost of inspection $c_1$, the cost of loss per unit time between failure and detection $c_2$, the cost of minimum repair of the detected breakdown of units $c_3$, and the overhaul cost $c_4$ of $R_s{\leq}R_g$. And in this paper, we will determine the optimal inspection policy to find the inspection period and number of tests that minimize the expected cost per unit time from the finite lifetime to the overhaul. Compare them through numerical examples.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.39 no.4
• /
• pp.381-387
• /
• 2002

In this paper, an efficient method to find shot boundaries in the MPEG video stream data is proposed. For this purpose, we first assume that the histogram difference value(HDV) and pixel difference value(PDV) as an one dimensional signal and apply the median filter to these signals. The output of the median filter is subtracted from the original signal to produce the median filtered difference(MFD). The MFD is a criterion of shot boundary. In addition a neural network is employed and trained to find exactly cut boundary. The proposed algorithm shows that the cut boundaries are well extracted, especially in a dynamic video.

• Journal of KIISE:Software and Applications
• /
• v.29 no.1_2
• /
• pp.114-125
• /
• 2002

In this paper, we are proposing a hierarchical segmentation method that first segments the video data into units of shots by detecting cut and dissolve, and then decides types of camera operations or object movements in each shot. In our previous work[1], each picture group is divided into one of the three detailed categories, Shot(in case of scene change), Move(in case of camera operation or object movement) and Static(in case of almost no change between images), by analysing DC(Direct Current) component of I(Intra) frame. In this process, we have designed two-stage hierarchical neural network with inputs of various multiple features combined. Then, the system detects the accurate shot position, types of camera operations or object movements by searching P(Predicted), B(Bi-directional) frames of the current picture group selectively and hierarchically. Also, the statistical distributions of macro block types in P or B frames are used for the accurate detection of cut position, and another neural network with inputs of macro block types and motion vectors method can reduce the processing time by using only DC coefficients of I frames without decoding and by searching P, B frames selectively and hierarchically. The proposed method classified the picture groups in the accuracy of 93.9-100.0% and the cuts in the accuracy of 96.1-100.0% with three different together is used to detect dissolve, types of camera operations and object movements. The proposed types of video data. Also, it classified the types of camera movements or object movements in the accuracy of 90.13% and 89.28% with two different types of video data.

• Proceedings of the IEEK Conference
• /
• 2000.06d
• /
• pp.7-10
• /
• 2000

There are many Problems such as low detection ratio, velocity and increase of false hit ratio on the detection of gradual scene changes with the previous shot transition detection algorithms. In this paper, we Propose an improved dissolve detection method using color information on low-frequency subband and edge elements on high-frequency subband. The Possible dissolve transition are found by analyzing the edge change ratio in the high-frequency subband with edge elements of each direction. Using the double chromatic difference on the lowest frequency subband, we have the improvement of the dissolve detection ratio. The simulation results show that the performance of the proposed algorithm is better than the conventional one for dissolve detection on a diverse set of uncompressed video sequences.

• The Journal of Korean Institute of Information Technology
• /
• v.17 no.5
• /
• pp.13-20
• /
• 2019

Recently, object detection technology based on CNN has been actively studied. Object detection technology is used as an important technology in autonomous vehicles, intelligent image analysis, and so on. In this paper, we propose a rotation change robust object detection system by applying MI-FL (Moment Invariant-Feature Layer) to SSD (Single Shot Multibox Detector) which is one of CNN-based object detectors. First, the features of the input image are extracted based on the VGG network. Then, a total of six feature layers are applied to generate bounding boxes by predicting the location and type of object. We then use the NMS algorithm to get the bounding box that is the most likely object. Once an object bounding box has been determined, the invariant moment feature of the corresponding region is extracted using MI-FL, and stored and learned in advance. In the detection process, it is possible to detect the rotated image more robust than the conventional method by using the previously stored moment invariant feature information. The performance improvement of about 4 ~ 5% was confirmed by comparing SSD with existing SSD and MI-FL.