• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.10C
• /
• pp.958-966
• /
• 2003

Change detection algorithms based on frame difference are frequently used for finding moving objects in image sequences. These algorithms detect the change of frames using estimated statistical background model. But, if this estimated background model is different from the actual statistical distribution, false detections are generated. In this paper, we propose an improved change detection algorithm using euler number and structure tensor. The proposed mapping method which is based on the euler number can be used for reducing the false detections that generated by nonparametric change detection algorithm. In this paper, the change in the region of moving object also can be detected by the proposed method using structure tensor. Experimental result shows that the proposed method reduces the false detections effectively by 90% on "Weather", by 34% on "Mother & daughter" and by 43% on "Aisle" than an existing method does.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2024.05a
• /
• pp.596-599
• /
• 2024

디지털 문서의 유통과정에서 발생할 수 있는 보안상의 문제를 해결하기 위해서는 파일 복사, 이동과정에 문서의 보안 등급을 자동 검출하고 특정 문서의 유출을 방지하는 보안 솔루션이 필요하다. 따라서 본 논문에서는 이러한 보안상의 문제를 해결하기 위하여 하나의 검출 분류 시스템을 제안하고자 한다. 제안한 시스템은 디지털 문서 내용을 이용하여 핵심 정보라고 판단되는 객체를 우선 추출한 후 그 핵심 유형을 분류하는 과정을 통해서 핵심 정보를 사전에 탐지하도록 하였다. 이를 위해서 SOTA를 달성한 YOLOv8를 이용하여 디지털 문서의 핵심 객체 감지하고 또한 파인튜닝을실시한 모델을 이용하여 그 유형을 분류하도록 설계하였다. 해당 시스템 검증을 위해서 기업에서 사용하고 있는 실제 사내 문서를 데이터셋을 이용하고 그 성능평가를 실시하였다.

• Journal of Broadcast Engineering
• /
• v.25 no.6
• /
• pp.944-953
• /
• 2020

Object detection plays a crucial role in a self-driving system. With the advances of image recognition based on deep convolutional neural networks, researches on object detection have been actively explored. In this paper, we proposed a lightweight model of the mask R-CNN, which has been most widely used for object detection, to efficiently predict location and shape of various objects on the road environment. Furthermore, feature maps are adaptively re-calibrated to improve the detection performance by applying an attention module to the neural network layer that plays different roles within the mask R-CNN. Various experimental results for real driving scenes demonstrate that the proposed method is able to maintain the high detection performance with significantly reduced network parameters.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2018.11a
• /
• pp.90-92
• /
• 2018

본 논문에서는 입력 영상을 카메라의 자세 정보에 따라 적절히 와핑한 후 이들을 심(Seam)을 따라 이어붙인 360VR 에서 갑작스런 객체 출현에 의해 중첩 영역에서 발생하는 왜곡 문제를 해결할 방법을 제안한다. 임의의 객체가 나타났을 때, 객체의 윤곽선을 반영하여 심(Seam)을 재설정함으로써 객체가 우그러지거나, 잘려나가는 등의 왜곡 문제를 해결한다. 이를 위해 본 논문에서는 가우시안(Gaussian) 혼합 모델 기반 전경/배경분리에 의한 움직이는 객체 추출, 객체 윤곽선 검출, 윤곽선에 기반한 심(Seam) 조정, 새로운 심(Seam) 기반 스티칭으로 왜곡을 없애는 방법을 제안하였다. 그리고 이를 실제 촬영 영상에 적용하여 왜곡 개선 효과를 보였다.

• Journal of KIISE:Software and Applications
• /
• v.34 no.1
• /
• pp.59-69
• /
• 2007

In the video with a various environment, background modeling is important for extraction and recognition the moving object. For this object recognition, many methods of the background modeling are proposed in a process of preprocess. Among these there is a Kumar method which represents the Queue-based background modeling. Because this has a fixed period of updating examination of the frame, there is a limit for various system. This paper use a background modeling based on the queue. We propose the method that major parameters are decided as adaptive by background model. They are the queue size of the sliding window, the sire of grouping by the brightness of the visual and the period of updating examination of the frame. In order to determine the factors, in every process, RCO (Ratio of Correct Object), REO (Ratio of Error Object) and UR (Update Ratio) are considered to be the standard of evaluation. The proposed method can improve the existing techniques of the background modeling which is unfit for the real-time processing and recognize the object more efficient.

• 전자공학회논문지 IE
• /
• v.43 no.2
• /
• pp.39-48
• /
• 2006

This paper is proposed the wavelet morphology algorithm with the spatial auto-object segmentation concept and the clustering concept. When it is segmented the color face by using the proposed algorithm, it is made to the simple image. Also, it is used the spatial quality in order to segment and detect the image as a real time without the user's manufacturing. This removed a small part that is regarded as a noise in image by HSV color model and applied the wavelet morphology to remove a part excepting for the face image. In this paper, it is made a comparison between the wavelet morphology algorithm and the morphology algorithm. And It is showed to accurately detect the face object parts in the image appled to HSV color space model.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.4
• /
• pp.851-859
• /
• 2011

In this paper, we propose the method that automatically extracts the silhouette and the joints of consecutive input image, and track joints to trace object for interaction between human and computer. Also the proposed method presents the action of human being to map human body using joints. To implement the algorithm, we model human body using 14 joints to refer to body size. The proposed method converts RGB color image acquired through a single camera to hue, saturation, value images and extracts body's silhouette using the difference between the background and input. Then we automatically extracts joints using the corner points of the extracted silhouette and the data of body's model. The motion of object is tracted by applying block-matching method to areas around joints among all image and the human's motion is mapped using positions of joints. The proposed method is applied to the test videos and the result shows that the proposed method automatically extracts joints and effectively maps human body by the detected joints. Also the human's action is aptly expressed to reflect locations of the joints

• Journal of Broadcast Engineering
• /
• v.25 no.3
• /
• pp.362-373
• /
• 2020

Recently, as autonomous vehicles and intelligent CCTV are growing more interest, the efficient object detection is essential technique. The DPM(Deformable Part Models) which is basis of this paper have used a typical object system that represents highly variable objects using mixtures of deformable part for object. Although it shows high detection performance by capturing part shape and configuration of object model, but it is limited to use in real application due to the complicated algorithm. In this paper, instead of image feature pyramid that takes up a large amount of computation in one part of the detector, we propose a method to reduce the computation speed by reconstructing a new image feature pyramid that uses adaptive bilinear interpolation of feature maps obtained on a specific image scale. As a result, the detection performance for object was lowered a little by 2.82%, however, the proposed detection method improved the speed performance by 10% in comparison with original DPM.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.3
• /
• pp.389-395
• /
• 2022

Because of improvement of deep learning techniques, deep learning using computer vision such as classification, detection and segmentation has also been used widely at many fields. Expecially, automatic driving is one of the major fields that applies computer vision systems. Also there are a lot of works and researches to combine multiple tasks in a single network. In this study, we propose the network that predicts the individual depth of pedestrians and vehicles. Proposed model is constructed based on YOLOv3 for object detection and Monodepth for depth estimation, and it process object detection and depth estimation consequently using encoder and decoder based on hard parameter sharing. We also used attention module to improve the accuracy of both object detection and depth estimation. Depth is predicted with monocular image, and is trained using self-supervised training method.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.5
• /
• pp.237-242
• /
• 2023

Research on high-performance walking robots is being actively conducted, and quadruped walking robots are receiving a lot of attention due to their excellent mobility and adaptability on uneven terrain, but they are difficult to introduce and utilize due to high cost. In this paper, to increase utilization by applying intelligent functions to a low-cost quadruped robot, we present a method of improving uneven terrain overcoming ability by mounting IMU and reinforcement learning on embedded board and automatically detecting objects using camera and deep learning. The robot consists of the legs of a quadruped mammal, and each leg has three degrees of freedom. We train complex terrain in simulation environments with designed 3D model and apply it to real robot. Through the application of this research method, it was confirmed that there was no significant difference in walking ability between flat and non-flat terrain, and the behavior of performing person detection in real time under limited experimental conditions was confirmed.