• 반도체디스플레이기술학회지
• /
• 제19권1호
• /
• pp.85-92
• /
• 2020

Object detection plays a critical role in the field of computer vision, and various researches have rapidly increased along with applying convolutional neural network and its modified structures since 2012. There are representative object detection algorithms, which are convolutional neural networks and YOLO. This paper presents two representative algorithm series, based on CNN and YOLO which solves the problem of CNN bounding box. We compare the performance of algorithm series in terms of accuracy, speed and cost. Compared with the latest advanced solution, YOLO v3 achieves a good trade-off between speed and accuracy.

• 한국멀티미디어학회논문지
• /
• 제24권6호
• /
• pp.796-804
• /
• 2021

In computer vision, visual tracking method addresses the problem of localizing an specific object in video sequence according to the bounding box. In this paper, we propose a tracking method by introducing the feature correlation comparison into the siamese network to increase its matching identification. We propose a way to compute location of object to improve matching performance by a correlation operation, which locates parts for solving the searching problem. The higher layer in the network can extract a lot of object information. The lower layer has many location information. To reduce error rate of the object center point, we built a siamese network that extracts the distribution and location information of target objects. As a result of the experiment, the average center error rate was less than 25%.

• 한국멀티미디어학회논문지
• /
• 제25권1호
• /
• pp.72-81
• /
• 2022

Visual Object Tracking is known as the most fundamental problem in the field of computer vision. Object tracking localize the region of target object with bounding box in the video. In this paper, a custom CNN is created to extract object feature that has strong and various information. This network was constructed as a Siamese network for use as a feature extractor. The input images are passed convolution block composed of a bottleneck layers, and features are emphasized. The feature map of the target object and the search area, extracted from the Siamese network, was input as a local proposal network. Estimate the object area using the feature map. The performance of the tracking algorithm was evaluated using the OTB2013 dataset. Success Plot and Precision Plot were used as evaluation matrix. As a result of the experiment, 0.611 in Success Plot and 0.831 in Precision Plot were achieved.

• 한국멀티미디어학회논문지
• /
• 제25권2호
• /
• pp.247-256
• /
• 2022

Visual tracking of selected target objects is fundamental challenging problems in computer vision. Object tracking localize the region of target object with bounding box in the video. We propose a Siam-FPN based custom fully CNN to solve visual tracking problems by regressing the target area in an end-to-end manner. A method of preserving the feature information flow using a feature map connection structure was applied. In this way, information is preserved and emphasized across the network. To regress object region and to classify object, the region proposal network was connected with the Siamese network. The performance of the tracking algorithm was evaluated using the OTB-100 dataset. Success Plot and Precision Plot were used as evaluation matrix. As a result of the experiment, 0.621 in Success Plot and 0.838 in Precision Plot were achieved.

• International Journal of Computer Science & Network Security
• /
• 제24권5호
• /
• pp.129-134
• /
• 2024

Object Detection is an emerging technology in the field of Computer Vision and Image Processing that deals with detecting objects of a particular class in digital images. It has considered being one of the complicated and challenging tasks in computer vision. Earlier several machine learning-based approaches like SIFT (Scale-invariant feature transform) and HOG (Histogram of oriented gradients) are widely used to classify objects in an image. These approaches use the Support vector machine for classification. The biggest challenges with these approaches are that they are computationally intensive for use in real-time applications, and these methods do not work well with massive datasets. To overcome these challenges, we implemented a Deep Learning based approach Convolutional Neural Network (CNN) in this paper. The Proposed approach provides accurate results in detecting objects in an image by the area of object highlighted in a Bounding Box along with its accuracy.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2021년도 춘계학술대회
• /
• pp.120-122
• /
• 2021

현재 국내외 자동차 수요가 증가하게 되면서 차량탑승 인원은 적어지고 차량 수는 증가하는 추세이다. 이는 교통체증이 더 심해지게 되는 주요 원인이 된다. 이를 해결하기 위해 다인승 전용차로, HOV(High-occupancy vehicle) lane을 운영하고 있지만, 이용 조건을 무시하고 불법으로 이용하는 사람들이 계속 증가하고 있다. 이러한 불법행위를 경찰이 육안으로 판단하여 단속하기 때문에 단속 정확도도 낮으며 효율이 떨어진다. 본 논문에서는 이와 같은 문제를 해결하기 위해 컴퓨터 비전을 이용한 영상 분석 기술을 이용해서 보다 효율적인 탐지를 할 수 있는 시스템 설계를 제안한다. 기존의 연구되었던 차량 탐지 방법을 개선하여 영상 안에서 트리거를 설정하여 탐지 객체가 선정된 후 대상에 대해서 집중적으로 영상 분석을 진행할 수 있게 설계했으며 딥러닝 객체 인식 모델인 YOLO 모델을 사용하여 실시간 객체 탐지와 정확한 신호를 얻기 위해 영상 내 bounding box로 판단하는 것이 아닌 중심점의 이동량을 이용하는 방법을 제안한다.

• 한국안전학회지
• /
• 제29권5호
• /
• pp.40-46
• /
• 2014

This paper deal with due to the additional equipment established on the vehicle fire accident cases. Recently, vehicle was recognized as other living spaces and it was increasing installation of additional devices such as a black box, multivision, additional battery and so on. However, these additional devices were established on the vehicle by non-specialists such as the general public user or manufacturing and distributors of additional devices since the vehicle factory. Resulting from the additional equipments due to installation errors in these additional equipments connected to the electrical wiring insulation damage or electrical breakdown occurs, and finally leading to a fire caused by the additional equipments. Therefore, additional equipment established on the vehicle due to a fire in order to prevent accidents during the manufacturing process of these devices are installed. If not installed unless the manufacturing process, the installation of additional equipments established by professional who knows the vehicle structure that will be able to prevent the fire.

• International Journal of Contents
• /
• 제14권2호
• /
• pp.24-29
• /
• 2018

In this paper, we propose an efficient traffic light detection (TLD) method as an advanced driver assistance system (ADAS) for elderly drivers. Since an increase in traffic accidents is associated with the aging population and an increase in elderly drivers causes a serious social problem, the provision of ADAS for older drivers via TLD is becoming a necessary(Ed: verify word choice: necessary?) public service. Therefore, we propose an economical TLD method that can be implemented with a simple black box (built in camera) and a smartphone in the near future. The system utilizes a color pre-processing method to differentiate between the stop and go signals. A mathematical morphology algorithm is used to further enhance the traffic light detection and a circular Hough transform is utilized to detect the traffic light correctly. From the simulation results of the computer vision and image processing based on a proposed algorithm on Matlab, we found that the proposed TLD method can detect the stop and go signals from the traffic lights not only in daytime, but also at night. In the future, it will be possible to reduce the traffic accident rate by recognizing the traffic signal and informing the elderly of how to drive by voice.

• 전기전자학회논문지
• /
• 제24권4호
• /
• pp.1141-1147
• /
• 2020

말벌 종은 모양이 매우 유사하기 때문에 비전문가가 분류하기 어렵고, 객체의 크기가 작고 빠르게 움직이기 때문에 실시간으로 탐지하여 종을 분류하는 것은 더욱 어렵다. 본 논문에서는 바운딩 박스를 이용한 딥러닝 알고리즘을 기반으로 말벌 종을 실시간으로 분류하는 시스템을 개발하였다. 훈련 영상의 레이블링 작업 시 바운딩 박스 안에 포함되는 배경 영역을 최소화하기 위하여 말벌의 머리와 몸통 부분만을 선택하는 방법을 제안한다. 또한 실시간으로 말벌을 탐지하고 그 종을 분류할 수 있는 최선의 알고리즘을 찾기 위하여 기존의 바운딩 박스 기반 객체 인식 알고리즘들을 실험을 통하여 비교한다. 실험 결과 컨볼루션 레이어의 활성함수로 mish 함수를 적용하고, 객체 검출 블록 전에 공간집중모듈(Spatial Attention Module, SAM)을 적용한 YOLOv4 모델을 사용하여 말벌 영상을 테스트한 경우 평균 97.89%의 정밀도(Precision)와 98.69%의 재현율(Recall)을 나타내었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권2호
• /
• pp.693-709
• /
• 2018

In this paper, we propose a compact Convolutional Neural Network (CNN)-based tracker in conjunction with a particle filter architecture, in which the CNN model operates as an accurate candidates estimator, while the particle filter predicts the target motion dynamics, lowering the overall number of calculations and refines the resulting target bounding box. Experiments were conducted on the Online Object Tracking Benchmark (OTB) [34] dataset and comparison analysis in respect to other state-of-art has been performed based on accuracy and precision, indicating that the proposed algorithm outperforms all state-of-the-art trackers included in the OTB dataset, specifically, TLD [16], MIL [1], SCM [36] and ASLA [15]. Also, a comprehensive speed performance analysis showed average frames per second (FPS) among the top-10 trackers from the OTB dataset [34].