• 한국기계가공학회지
• /
• 제20권12호
• /
• pp.71-77
• /
• 2021

An impeller identification technique that is essential for adding viscosity measurement functions to overhead stirrers is presented in this study. Previous studies have revealed that using magnets facing the same poles arranged in a row can aid in distinguishing the types of impellers by detecting the number of magnets in a non-contact manner. However, as these previous studies measured the magnetic fields using analog Hall sensors, a converting circuit for the digital signals is required that can interface with the MCU. In this study, it was demonstrated that the number of magnets can be distinguished without using a separate conversion circuit by using a Hall sensor with a digital output. Owing to the unique hysteresis characteristics of digital Hall sensors, it was confirmed through experiments that the complex and diverse outputs appear depending on the direction of the magnetic field, the arrangement of magnetic poles, and the moving direction of the magnet. The measurement of the magnetic field showed that an edge signal equal to the number of magnets inserted into the impeller was detected when the radial direction was used, and the south pole was first approached.

• 한국기계가공학회지
• /
• 제18권7호
• /
• pp.56-62
• /
• 2019

The high-precision laser scriber carries out scribing alumina ceramic substrates for manufacturing ultra-small chip resistors. The ceramic substrates are loaded, aligned, scribed, transferred, and unloaded. The entire process is fully automated, thereby minimizing the scribing cycle time of the ceramic substrates and improving the throughput. The scriber consists of the laser optical system, pick-up module of ceramic substrates, pre-alignment module, TH axis drive work table, automation module for substrate loading / unloading, and high-speed scribing control S/W. The loader / unloader unit, which has the greatest influence on the scribing cycle time of the substrates, carries the substrates to the work table that carries out the cutting line work by driving the X and Y axes as well as by adsorbing the ceramic substrates. The loader / unloader unit consists of the magazine up / down part, X-axis drive part for conveying the substrates to the left and right direction, and the vision part for detecting the edge of the substrate for the primary pre-alignment of the substrates. In this paper, the laser scribing machining simulation is performed by applying the instrument mechanism of each component module. Through this study, the scribing machining process is first verified by analyzing the process operation and work area of each module in advance. In addition, the scribing machining process is optimized by comparing and analyzing the scribing cycle time of one ceramic substrate according to the alignment stage module speed.

• 한국정보기술학회 영문논문지
• /
• 제9권1호
• /
• pp.115-125
• /
• 2019

In this paper, the image is received from the camera and the lane is sensed. There are various ways to detect lanes. Generally, the method of detecting edges uses a lot of the Sobel edge detection and the Canny edge detection. The minimum use of multiplication and division is used when designing for the hardware configuration. The images are tested using a black box image mounted on the vehicle. Because the top of the image of the used the black box is mostly background, the calculation process is excluded. Also, to speed up, YCbCr is calculated from the image and only the data for the desired color, white and yellow lane, is obtained to detect the lane. The median filter is used to remove noise from images. Intermediate filters excel at noise rejection, but they generally take a long time to compare all values. In this paper, by using addition, the time can be shortened by obtaining and using the result value of the median filter. In case of the Sobel edge detection, the speed is faster and noise sensitive compared to the Canny edge detection. These shortcomings are constructed using complementary algorithms. It also organizes and processes data into parallel processing pipelines. To reduce the size of memory, the system does not use memory to store all data at each step, but stores it using four line buffers. Three line buffers perform mask operations, and one line buffer stores new data at the same time as the operation. Through this work, memory can use six times faster the processing speed and about 33% greater quantity than other methods presented in this paper. The target operating frequency is designed so that the system operates at 50MHz. It is possible to use 2157fps for the images of 640by360 size based on the target operating frequency, 540fps for the HD images and 240fps for the Full HD images, which can be used for most images with 30fps as well as 60fps for the images with 60fps. The maximum operating frequency can be used for larger amounts of the frame processing.

• 센서학회지
• /
• 제12권2호
• /
• pp.57-65
• /
• 2003

CMOS 공정을 이용한 윤곽검출 시각칩 설계시, 넓은 범위의 광강도에 대해서 이미지의 특징검출을 위하여 국소 광적응기능이 필요하다. 국소 광적응이란 망막내 수평억제(lateral inhibition) 기능을 행하는 수평세포를 이용하여 입력 광강도에 응답하는 국소적인 수평세포층의 수용야 크기를 변화시켜 동일한 출력레벨을 얻는 것이다. 따라서, 배경광보다 조금 크거나 아주 큰 입력광의 변화가 있을 때 동일한 출력레벨을 얻을 수 있다. 본 연구에서는, 망막내 수평세포를 p-MOSFET로 구성된 저항성 회로망으로 모델링 및 해석하고, 이를 이용하여 설계된 시각칩의 국소 광적응 메커니즘을 검증하였다.

• 대한전자공학회논문지SP
• /
• 제47권6호
• /
• pp.19-27
• /
• 2010

본 논문은 B-스플라인 능동적 윤곽을 차 에지 영상에 적용하여 얼굴을 검출함에 있어, 검출 결과의 정확도를 제고하고 연산량을 감소시키는 방법을 제안한다. 제안 방법은 먼저, 차이 영상의 첨도(kurtosis)를 이용하여 사용자의 움직임량을 추정한다. 이때, 첨도 값에 따라 사용자의 움직임량이 작다고 판단된 경우에는 윤곽선 적합을 실시하지 않으며, 움직임량이 크다고 판단된 경우에만 윤곽선 적합을 실시한다. 그 후, 윤곽선 적합을 위하여 이진화된 차이 영상의 거리변환(distance transform)된 결과와 현재 영상의 에지(edge)를 사용하여 움직임과 관련된 차 에지 영상을 추출하고, 마지막으로 이렇게 추출된 차 에지 영상에 윤곽선 적합을 실시하여 얼굴의 위치를 검출하게 된다. 첨도를 이용하여 사용자의 움직임량을 추정하는 방법은 윤곽선 적합 결과를 안정화시켜주는 동시에 연산량을 절약시켜주며, 현재 영상의 에지와 이진화된 차이 영상의 거리변환을 사용한 움직임 에지 추정 방법은 윤곽선 처짐과 불연속적인 에지 추출의 문제점을 개선시켜준다. 실험을 통해, 제안한 방법이 기존의 윤곽선 처짐이나 에지 끊어짐에 의한 오류를 줄여 주는 동시에, 약 39%의 영상에 대한 윤곽선 적합을 생략시켜주어 연산량을 줄여 줄 수 있음을 확인하였다.

• 국제학술발표논문집
• /
• The 9th International Conference on Construction Engineering and Project Management
• /
• pp.1243-1244
• /
• 2022

In recent years, the growing interest in off-site construction has led to factories scaling up their manufacturing and production processes in the construction sector. Consequently, continuous large-scale site monitoring in low-variability environments, such as prefabricated components production plants (precast concrete production), has gained increasing importance. Although many studies on computer vision-based site monitoring have been conducted, challenges for deploying this technology for large-scale field applications still remain. One of the issues is collecting and transmitting vast amounts of video data. Continuous site monitoring systems are based on real-time video data collection and analysis, which requires excessive computational resources and network traffic. In addition, it is difficult to integrate various object information with different sizes and scales into a single scene. Various sizes and types of objects (e.g., workers, heavy equipment, and materials) exist in a plant production environment, and these objects should be detected simultaneously for effective site monitoring. However, with the existing object detection algorithms, it is difficult to simultaneously detect objects with significant differences in size because collecting and training massive amounts of object image data with various scales is necessary. This study thus developed a large-scale site monitoring system using edge computing and a small-object detection system to solve these problems. Edge computing is a distributed information technology architecture wherein the image or video data is processed near the originating source, not on a centralized server or cloud. By inferring information from the AI computing module equipped with CCTVs and communicating only the processed information with the server, it is possible to reduce excessive network traffic. Small-object detection is an innovative method to detect different-sized objects by cropping the raw image and setting the appropriate number of rows and columns for image splitting based on the target object size. This enables the detection of small objects from cropped and magnified images. The detected small objects can then be expressed in the original image. In the inference process, this study used the YOLO-v5 algorithm, known for its fast processing speed and widely used for real-time object detection. This method could effectively detect large and even small objects that were difficult to detect with the existing object detection algorithms. When the large-scale site monitoring system was tested, it performed well in detecting small objects, such as workers in a large-scale view of construction sites, which were inaccurately detected by the existing algorithms. Our next goal is to incorporate various safety monitoring and risk analysis algorithms into this system, such as collision risk estimation, based on the time-to-collision concept, enabling the optimization of safety routes by accumulating workers' paths and inferring the risky areas based on workers' trajectory patterns. Through such developments, this continuous large-scale site monitoring system can guide a construction plant's safety management system more effectively.

• 마이크로전자및패키징학회지
• /
• 제22권1호
• /
• pp.15-19
• /
• 2015

본 연구에서는 OSP (organic solderability preservative) 표면처리된 PCB (printed circuit board) Cu 볼 패드의 변색을 검사하는 측정 시스템을 제안하였다. PCB 표면처리 중에서 OSP는 친환경적, 낮은 생산 비용 등의 장점으로 널리 이용되고 있으나 온도공정에 따른 변색이 발생하는 문제점이 있어서 접합 신뢰성 불량의 한 원인이 되고 있다. 이러한 변색 불량을 장치 비의존적 CIELAB 색좌표를 도입하여 분석하였다. 먼저, PCB 샘플을 검사하기 위해 적합한 측정 시스템을 표준 조명과 CCD 카메라를 이용하여 제작하고, 랩뷰 (labview) 프로그램을 이용하여 Cu 볼 패드의 변색을 검사하기 위한 이미지를 얻는 알고리즘을 제작하였다. 전체 PCB 이미지에서 이진화 (binarization) 및 외곽영역 추적 (edge detection) 영상처리 과정을 통하여 Cu 볼 패드만의 이미지를 획득하고, 장치 의존적인 RGB 색좌표에서 $3{\times}3$ 변환 행렬을 이용하여 CIELAB 색좌표로 변환하는 과정을 거친다. 본 측정 시스템을 이용하여 변색이 발생한 PCB 샘플을 분석한 결과 Cu 볼 패드 만의 이미지를 대상으로 분석하면 연산에 소요되는 시간이 감소하고 측정 시스템의 오인식률을 감소 시킬 수 있음을 실험적으로 증명하였다. 또한 CIELAB 색좌표 중 $L^*$ (밝음-어두움의 정도), $b^*$ (노랑-파랑의 정도)의 두 가지 기준의 조합이 Cu 볼 패드의 변색 검사에 적합한 색좌표로 분석되었다.

• Journal of Multimedia Information System
• /
• 제3권4호
• /
• pp.119-128
• /
• 2016

This study aims to suggest a new algorithm for detecting two-wheelers on road that have various shapes according to the viewing angle for vision based intelligent vehicles. This article describes a new approach to two-wheelers detection algorithm riding on people based on modified Histogram of Oriented Gradients (HOG) using correlation coefficient (CC). The CC between two local area variables, in which one is the person riding a bike and other is its background, can represent correlation relation. First, we extract edge vectors using HOG which includes gradient information and differential magnitude as cell based. And then, the value, which is calculated by the CC between the area of each cell and one of two-wheelers, can be extracted as the weighting factor in process for normalizing the modified HOG cell. This paper applied the Adaboost algorithm to make a strong classification from weak classification. In this experiment, we can get the result that the detection rate of the proposed method is higher than that of the traditional method.

• 전자공학회논문지B
• /
• 제32B권1호
• /
• pp.127-135
• /
• 1995

The diffusion neural network forms a Gaussian distribution by transferring an excitation to the surround. A DOG(difference of two Gaussians) is obtained by the diffusion neural network. This type of the DOG, which can detect the intensity changes of an image, has the same shape as a LOG(Laplacian of a Gaussian:${\Delta}^2$G) and narrow band pass characteristics. In this paper we show that another type of the DOG which has a very narrow Gaussian for the excitatory and a very wide Gaussian for the inhibitory, can be formed by the diffusion process of this network, This type of the DOG has a wide band width in spatial frequency domain and can be used efficiently in detecting special type of edges.

• Journal of Information Processing Systems
• /
• 제13권1호
• /
• pp.128-140
• /
• 2017

The wireless sensor networks (WSNs) became a very essential tool in borders and military zones surveillance, for this reason specific applications have been developed. Surveillance is usually accomplished through the deployment of nodes in a random way providing heterogeneous topologies. However, the process of the identification of all nodes located on the network's outer edge is very long and energy-consuming. Before any other activities on such sensitive networks, we have to identify the border nodes by means of specific algorithms. In this paper, a solution is proposed to solve the problem of energy and time consumption in detecting border nodes by means of node selection. This mechanism is designed with several starter nodes in order to reduce time, number of exchanged packets and then, energy consumption. This method consists of three phases: the first one is to detect triggers which serve to start the mechanism of boundary nodes (BNs) detection, the second is to detect the whole border, and the third is to exclude each BN from the routing tables of all its neighbors so that it cannot be used for the routing.