• IEIE Transactions on Smart Processing and Computing
• /
• v.4 no.2
• /
• pp.110-114
• /
• 2015

This paper reports a novel object-detection technique in video sequences. The proposed algorithm consists of detection of objects in illuminated and occluded videos by using object features and a neural network technique. It consists of two functional modules: region-based object feature extraction and continuous detection of objects in video sequences with region features. This scheme is proposed as an enhancement of the Lowe's scale-invariant feature transform (SIFT) object detection method. This technique solved the high computation time problem of feature generation in the SIFT method. The improvement is achieved by region-based feature classification in the objects to be detected; optimal neural network-based feature reduction is presented in order to reduce the object region feature dataset with winner pixel estimation between the video frames of the video sequence. Simulation results show that the proposed scheme achieves better overall performance than other object detection techniques, and region-based feature detection is faster in comparison to other recent techniques.

• Journal of the Korea Society of Computer and Information
• /
• v.24 no.8
• /
• pp.19-28
• /
• 2019

In this paper, an online video summarization algorithm based on multiple object detection is proposed. As crime has been on the rise due to the recent rapid urbanization, the people's appetite for safety has been growing and the installation of surveillance cameras such as a closed-circuit television(CCTV) has been increasing in many cities. However, it takes a lot of time and labor to retrieve and analyze a huge amount of video data from numerous CCTVs. As a result, there is an increasing demand for intelligent video recognition systems that can automatically detect and summarize various events occurring on CCTVs. Video summarization is a method of generating synopsis video of a long time original video so that users can watch it in a short time. The proposed video summarization method can be divided into two stages. The object extraction step detects a specific object in the video and extracts a specific object desired by the user. The video summary step creates a final synopsis video based on the objects extracted in the previous object extraction step. While the existed methods do not consider the interaction between objects from the original video when generating the synopsis video, in the proposed method, new object clustering algorithm can effectively maintain interaction between objects in original video in synopsis video. This paper also proposed an online optimization method that can efficiently summarize the large number of objects appearing in long-time videos. Finally, Experimental results show that the performance of the proposed method is superior to that of the existing video synopsis algorithm.

• International Journal of Internet, Broadcasting and Communication
• /
• v.11 no.4
• /
• pp.76-85
• /
• 2019

We introduce a vision-based object detection method for real-time video surveillance system in low-end edge computing environments. Recently, the accuracy of object detection has been improved due to the performance of approaches based on deep learning algorithm such as Region Convolutional Neural Network(R-CNN) which has two stage for inferencing. On the other hand, one stage detection algorithms such as single-shot detection (SSD) and you only look once (YOLO) have been developed at the expense of some accuracy and can be used for real-time systems. However, high-performance hardware such as General-Purpose computing on Graphics Processing Unit(GPGPU) is required to still achieve excellent object detection performance and speed. To address hardware requirement that is burdensome to low-end edge computing environments, We propose sub-frame analysis method for the object detection. In specific, We divide a whole image frame into smaller ones then inference them on Convolutional Neural Network (CNN) based image detection network, which is much faster than conventional network designed forfull frame image. We reduced its computationalrequirementsignificantly without losing throughput and object detection accuracy with the proposed method.

• Journal of Korea Multimedia Society
• /
• v.24 no.7
• /
• pp.890-902
• /
• 2021

Although the object detection accuracy with still images has been significantly improved with the advance of deep learning techniques, the object detection problem with video data remains as a challenging problem due to the real-time requirement and accuracy drop with occlusion. In this research, we propose a method in pig detection for video monitoring environment. First, we determine a motion, from a video data obtained from a tilted-down-view camera, based on the average size of each pig at each location with the training data, and extract key frames based on the motion information. For each key frame, we then apply YOLO, which is known to have a superior trade-off between accuracy and execution speed among many deep learning-based object detectors, in order to get pig's bounding boxes. Finally, we merge the bounding boxes between consecutive key frames in order to reduce false positive and negative cases. Based on the experiment results with a video data set obtained from a pig farm, we confirmed that the pigs could be detected with an accuracy of 97% at a processing speed of 37fps.

• Journal of Information Processing Systems
• /
• v.19 no.4
• /
• pp.439-449
• /
• 2023

This paper deals with a resource efficient artificial intelligence (AI) service architecture for multi-channel video streams. As an AI service, we consider the object detection model, which is the most representative for video applications. Since most object detection models are basically designed for a single channel video stream, the utilization of the additional resource for multi-channel video stream processing is inevitable. Therefore, we propose a resource efficient AI service framework, which can be associated with various AI service models. Our framework is designed based on the modular architecture, which consists of adaptive frame control (AFC) Manager, multiplexer (MUX), adaptive channel selector (ACS), and YOLO interface units. In order to run only a single YOLO process without regard to the number of channels, we propose a novel approach efficiently dealing with multi-channel input streams. Through the experiment, it is shown that the framework is capable of performing object detection service with minimum resource utilization even in the circumstance of multi-channel streams. In addition, each service can be guaranteed within a deadline.

• Journal of Information Processing Systems
• /
• v.17 no.2
• /
• pp.227-241
• /
• 2021

Object detection techniques based on deep learning such as YOLO have high detection performance and precision in a single channel video stream. In order to expand to multiple channel object detection in real-time, however, high-performance hardware is required. In this paper, we propose a novel back-end server framework, a real-time AI vision platform (RAVIP), which can extend the object detection function from single channel to simultaneous multi-channels, which can work well even in low-end server hardware. RAVIP assembles appropriate component modules from the RODEM (real-time object detection module) Base to create per-channel instances for each channel, enabling efficient parallelization of object detection instances on limited hardware resources through continuous monitoring with respect to resource utilization. Through practical experiments, RAVIP shows that it is possible to optimize CPU, GPU, and memory utilization while performing object detection service in a multi-channel situation. In addition, it has been proven that RAVIP can provide object detection services with 25 FPS for all 16 channels at the same time.

• Journal of Intelligence and Information Systems
• /
• v.28 no.1
• /
• pp.89-106
• /
• 2022

Recently, the amount of video data collected from smartphones, CCTVs, black boxes, and high-definition cameras has increased rapidly. According to the increasing video data, the requirements for analysis and utilization are increasing. Due to the lack of skilled manpower to analyze videos in many industries, machine learning and artificial intelligence are actively used to assist manpower. In this situation, the demand for various computer vision technologies such as object detection and tracking, action detection, emotion detection, and Re-ID also increased rapidly. However, the object detection and tracking technology has many difficulties that degrade performance, such as re-appearance after the object's departure from the video recording location, and occlusion. Accordingly, action and emotion detection models based on object detection and tracking models also have difficulties in extracting data for each object. In addition, deep learning architectures consist of various models suffer from performance degradation due to bottlenects and lack of optimization. In this study, we propose an video analysis system consists of YOLOv5 based DeepSORT object tracking model, SlowFast based action recognition model, Torchreid based Re-ID model, and AWS Rekognition which is emotion recognition service. Proposed model uses single-linkage hierarchical clustering based Re-ID and some processing method which maximize hardware throughput. It has higher accuracy than the performance of the re-identification model using simple metrics, near real-time processing performance, and prevents tracking failure due to object departure and re-emergence, occlusion, etc. By continuously linking the action and facial emotion detection results of each object to the same object, it is possible to efficiently analyze videos. The re-identification model extracts a feature vector from the bounding box of object image detected by the object tracking model for each frame, and applies the single-linkage hierarchical clustering from the past frame using the extracted feature vectors to identify the same object that failed to track. Through the above process, it is possible to re-track the same object that has failed to tracking in the case of re-appearance or occlusion after leaving the video location. As a result, action and facial emotion detection results of the newly recognized object due to the tracking fails can be linked to those of the object that appeared in the past. On the other hand, as a way to improve processing performance, we introduce Bounding Box Queue by Object and Feature Queue method that can reduce RAM memory requirements while maximizing GPU memory throughput. Also we introduce the IoF(Intersection over Face) algorithm that allows facial emotion recognized through AWS Rekognition to be linked with object tracking information. The academic significance of this study is that the two-stage re-identification model can have real-time performance even in a high-cost environment that performs action and facial emotion detection according to processing techniques without reducing the accuracy by using simple metrics to achieve real-time performance. The practical implication of this study is that in various industrial fields that require action and facial emotion detection but have many difficulties due to the fails in object tracking can analyze videos effectively through proposed model. Proposed model which has high accuracy of retrace and processing performance can be used in various fields such as intelligent monitoring, observation services and behavioral or psychological analysis services where the integration of tracking information and extracted metadata creates greate industrial and business value. In the future, in order to measure the object tracking performance more precisely, there is a need to conduct an experiment using the MOT Challenge dataset, which is data used by many international conferences. We will investigate the problem that the IoF algorithm cannot solve to develop an additional complementary algorithm. In addition, we plan to conduct additional research to apply this model to various fields' dataset related to intelligent video analysis.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.723-726
• /
• 2021

현재 단일 이미지에서 Object Detection 성능은 매우 좋은 편이다. 하지만 동영상에서는 처리 속도가 너무 느리고 임베디드 시스템에서는 real-time이 힘든 상황이다. 연구 논문에서는 하이엔드 GPU에서 다른 기능 없이 YOLO만 구동했을 때 real-time이 가능하다고 하지만 실제 사용자들은 상대적으로 낮은 사양의 GPU를 사용하거나 CPU를 사용하기 때문에 일반적으로는 자연스러운 real-time을 하기가 힘들다. 본 논문에서는 이러한 제한점을 해결하고자 계산량이 많은 Object Detection model 사용을 줄이는 방안은 제시하였다. 현재 Video영상에서 Object Detection을 수행할 때 매 frame마다 YOLO모델을 구동하는 것에서 YOLO 사용을 줄임으로써 계산 효율을 높였다. 본 논문의 알고리즘은 카메라가 움직이거나 배경이 바뀌는 상황에서도 사용이 가능하다. 속도는 최소2배에서 ~10배이상까지 개선되었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.2
• /
• pp.771-788
• /
• 2020

Detection of abandoned objects for smart video surveillance should be robust and accurate in various situations with low computational costs. This paper presents a new algorithm for abandoned object detection based on the dual background model. Through the template registration of a candidate stationary object and presence authentication methods presented in this paper, we can handle some complex cases such as occlusions, illumination changes, long-term abandonment, and owner's re-attendance as well as general detection of abandoned objects. The proposed algorithm also analyzes video frames at specific intervals rather than consecutive video frames to reduce the computational overhead. For performance evaluation, we experimented with the algorithm using the well-known PETS2006, ABODA datasets, and our video dataset in a live streaming environment, which shows that the proposed algorithm works well in various situations.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.574-578
• /
• 2009

Real-time object detection for distinguishing a moving object of interests from the background image in still image or video image sequence is an essential step to a correct object tracking and recognition. Moving cast shadow can be misclassified as part of objects or moving objects because the shadow region is included in the moving object region after object segmentation. For this reason, an algorithm for shadow removal plays an important role in the results of accurate moving object detection and tracking systems. To handle with the problems, an accurate algorithm based on the features of moving object and shadow in color space is presented in this paper. Experimental results show that the proposed algorithm is effective to detect a moving object and to remove shadow in test video sequences.