• The Korean Journal of Applied Statistics
• /
• v.30 no.3
• /
• pp.487-501
• /
• 2017

A machine vision system (MVS) is a computer system that utilizes one or more image-capturing devices to provide image data for analysis and interpretation. Recently there have been a number of industrial- and medical-device applications where control charts have been proposed for use with image data. The use of image-based control charting is somewhat different from traditional control charting applications, and these differences can be attributed to several factors, such as the type of data monitored and how the control charts are applied. In this paper, we investigate the adjustment effect of image size and region of interest (ROI) size, when we use control charts to monitor grayscale image data in industry.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.35 no.6
• /
• pp.599-608
• /
• 2017

Cracks on the asphalt road surface can affect the speed of the car, the consumption of fuel, the ride quality of the road, and the durability of the road surface. Such cracks in roads can lead to very dangerous consequences for long periods of time. To prevent such risks, it is necessary to identify cracks and take appropriate action. It takes too much time and money to do it. Also, it is difficult to use expensive laser equipment vehicles for initial cost and equipment operation. In this paper, we propose an effective detection method of road surface defect using ROI (Region of Interest) setting and cany edge detection method using UAV image. The results of this study can be presented as efficient method for road surface flaw detection and maintenance using UAV. In addition, it can be used to detect cracks such as various buildings and civil engineering structures such as buildings, outer walls, large-scale storage tanks other than roads, and cost reduction effect can be expected.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.311-312
• /
• 2022

본 논문에서는 딥러닝 기반 욕창 감지를 위한 욕창 객체 탐지를 연구한다. 객체 탐지 딥러닝 기법으로 RCNN, Fast R-CNN, Faster R-CNN, YOLO 등 다양한 기법이 존재하며, 각 모델의 특징 또한 다르다. 욕창은 단계별로 피부, 조직에 손상의 정도가 다르다. 낮은 단계의 경우 일반적인 피부색과 유사하게 나타나며, 높은 단계의 경우 근육, 뼈, 지지 조직 등의 괴사로 인해 삼출물 또는 괴사조직이 나타난다. 논문에서는 One-Stage Detection 기법인 YOLO를 기반으로 욕창 이미지 내부에서 욕창 탐지를 진행한다. 현재 보유하고 있는 이미지 데이터 수가 많지 않아 데이터 증강기법을 통해 데이터를 증강하여 학습에 활용하였다.

• The KIPS Transactions:PartB
• /
• v.15B no.3
• /
• pp.189-196
• /
• 2008

In this paper, a CMOS digital image sensor, which consists of A/D conversion, motion estimation circuits, and an attention module for ROI (Region of Interest) detection is presented. The functions of A/D conversion and motion estimation are implemented by $0.6{\mu}m$ CMOS processing circuit as hardware, and the attention module is implemented outside the circuit as software currently. Attention modules are taken to improve limited applications of the smart image sensor. The current smart image sensor responses to the changes of intensity, and uses the integration time to estimate motion. Therefore it is limited in its applications. To make up for inherent property of the sensor from circuit design and extend its applications we decide to introduce perception solutions to the image sensor. Attention modules for still and moving images are employed to achieve such purposes. The suggested approach makes the smart image sensor available with additional functions for such cases that motion estimation or intensity changes are not observed. Experimental result shows the usefulness and extension of the image sensor.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.11
• /
• pp.1437-1444
• /
• 2020

With the development of computer vision systems, many advances have been made in the fields of surveillance, biometrics, medical imaging, and autonomous driving. In the field of autonomous driving, in particular, the object detection technique using deep learning are widely used, and the paved road detection is a particularly crucial problem. Unlike the ROI detection algorithm used in general object detection, the structure of paved road in the image is heterogeneous, so the ROI-based object recognition architecture is not available. In this paper, we propose a deep neural network architecture for atypical paved road detection using Semantic segmentation network. In addition, we introduce the multi-scale semantic segmentation network, which is a network architecture specialized to the paved road detection. We demonstrate that the performance is significantly improved by the proposed method.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.26 no.6
• /
• pp.493-497
• /
• 2016

Infrared (IR) imaging is capable of detecting targets that are not visible at night, thus it has been widely used for the security and defense system. However, the quality of the IR image is often degraded by low resolution and noise corruption. This paper addresses target segmentation with the IR image. Multiple regions of interest (ROI) are extracted by the multi-level segmentation and targets are segmented from the individual ROI. Each level of the multi-level segmentation is composed of a k-means clustering algorithm an expectation-maximization (EM) algorithm, and a decision process. The k-means clustering algorithm initializes the parameters of the Gaussian mixture model (GMM) and the EM algorithm iteratively estimates those parameters. Each pixel is assigned to one of clusters during the decision. This paper proposes the selection and the merging of the extracted ROIs. ROI regions are selectively merged in order to include the overlapped ROI windows. In the experiments, the proposed method is tested on an IR image capturing two pedestrians at night. The performance is compared with conventional methods showing that the proposed method outperforms others.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.10a
• /
• pp.1103-1105
• /
• 2015

Leaked Radioactive source in nuclear power station, radiation related facilities and the aging nuclear power plant for the dismantling must need to detect and remove early to prevent major accidents. In this paper, we implemented a single sensor-based gamma-ray detectors stereo which can provide the distance to the radiation source, a direction and doserate information for fast and efficient decontamination work the radiation source. And we have carried out an algorithm development for high-speed detection of the detection equipment. Two detectors are required for stereo structure for obtaining the distance information of the radioactive source, but we designed the only sensor-based detection device for the weight reduction. We have extracted the region of interest and obtained the distance calculation result and distribution of radiation source in order to minimize a stereo image acquisition time. Detection time of the algorithm showed a shorter time of about 41%.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2016.05a
• /
• pp.780-781
• /
• 2016

The stereo radiation detection system detects gamma ray source and measures the two dimensional distribution image based on the detection result. Then the system is implemented to measure the distance to the radiation source from the system in 3D space using stereo vision algorithm. In this paper, we reduced the time for a gamma-ray scan space detection through image processing algorithms. In addition, it combines radiation and visible light images. Then we conducted a study for improving the distribution of gamma-ray detection efficiency through the stereo calibration using a 3D visualization. As a result, we obtain an improved detection time by more than 30% and have acquired a visible image with a 3D monitor.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.10 no.6
• /
• pp.55-62
• /
• 2010

Monitoring systems are able to report certain events at region of interest(ROI) and to take an appropriate action. From industrial product line full of robots to fire detection, intrusion detection, smart grid application, environmental pollution alarm system, monitoring system has widely used in diverse industry sector. Recently, due to advance of wireless communication technology and availability of low cost sensors, intelligent and/or smart monitoring systems such as sensor networks has been developed. Several deployment methods are introduced to meet various monitoring needs and deployment performance criteria are also summarized to be used to identify weak point and be useful at designing monitoring systems. Both efficiency during deployment and usefulness after the deployment should be assessed. Efficiency factors during deployment are elapsed time, energy required, deployment cost, safety, sensor node failure rate, scalability. Usefulness factors after deployment are ROI coverage, connectivity, uniformity, target density similarity, energy consumption rate per unit time and so on.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04c
• /
• pp.316-318
• /
• 2003

간질성 폐질환(Interstitial Lung Disease) 컴퓨터지원진단(Computer-Aided Diagnosis： CAD)시스템은 방사선의사들이 흉부 X-ray영상에서 석회화와 섬유화를 탐지하고자 적용하였다. 진단 중에 발생할 수 있는 오진율을 줄이고 간질성 폐질환이 존재하는 폐야에서 이상유무를 판단하여 검출을 표시하도록 하였다. 본 논문에서는 디지털 흉부영상에서의 간질성 폐질환을 검출하기 위해 폐 텍스처(texture)의 물리적 척도를 측정하기 위한 방법을 제안한다. 2차원의 푸리에 변환으로부터 얻어지는 파워스펙트럼(power spectrum) 분석에 기반을 두는 방법으로 각각의 ROI(Region Of Interest)에서 구한 평균제곱자승오차(Root Mean Sguare: RMS)와 파워스펙트럼의 첫 번갠 모멘트(Moment)는 폐 텍스처의 밀도변동의 크기(magnitude)와 섬세함(fineness)을 나타낸다. 실험결과 다양한 간질성폐질환을 가진 비정상 폐 텍스처의 RMS와 첫 번째 모멘트와는 차이가 있었다. 디지텔 흉부영상으로부터 계산되어진 정량화된 텍스처의 척도는 방사선의사의 간질성 폐 질환을 진단함에 효율적인 질환 탐지를 가능하게 하였으며 진단율을 향상시킬 수 있었다.