• 대한의용생체공학회:의공학회지
• /
• 제12권2호
• /
• pp.89-98
• /
• 1991

This study is concerned on the automation for cell diagnosis which has better objectivity and speed of test than human beings. Diagnosis is on the basis of shape change of abnormal Cells. Used parameters are nucleus area, nucleus perimeter, nucleus shape, cytoplasm area, nucleus/cytoplsm ratio, which was obtained using image processing technics. A new mode method is proposed on the automatic threshold selection for superior process time compared with Otsu's. Contour of the cytoplasm of abnormal cell is obtained using me- dian filter and sorel operator. The mask to get only original shape of abnormal cells is formed uslng the contour filling algorithm. In the result the normal cells are separated from the abnormal cells and the abnormal cells can be distinguished through screwing of abnormal cell's image with reference data to judge abnormal cells. Owing to this study the number of inspections which the pathologists should examine will be decreased and the time for inspection will be shortened.

• 제어로봇시스템학회논문지
• /
• 제17권5호
• /
• pp.479-483
• /
• 2011

The automation of production process is actively expanding for the purpose of the cost reduction and quality assurance. Among these, automatic tracking of the product along the whole process of the production is also important topic. Typically this is done by adopting OCR technology. Conventional OCR technology operates well on the rather good quality of the image like as printed characters on the paper. In industrial application, IDs are marked on the metal surface, and this cause the height difference between background material and character. Illumination systems that guarantee an image with good quality may be a solution, but it is rather difficult to design such an illumination system. This paper proposes an algorithm for the recognition of vehicle's ID characters using edge projection and PCA (Principal Component Analysis). Proposed algorithm robustly operates under illumination change using the same parameters. Experimental results show the feasibility of the proposed algorithm.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2009년도 IWAIT
• /
• pp.504-507
• /
• 2009

Color chart area is automatically extracted in image that captured a crop such as fruits with the color chart, and an approximation formula is obtained for the change in feature value of the color indexes. Comparison is made with the color value of the crop area, and the growing degree is assessed according to the correlation. Using a compact PC equipped with the program, image of fruits is captured, and the output value obtained by the system is compared to the rating by expert. In the automatic recognition of the color chart out of doors, the complete color indexes is correctly acquired in 22 of 29 images. And indoors, they are correctly acquired in all of 34 images. In the color value judgment of the Japanese pear, indoors, 32 of 34 images is within 1.0 of the judgment error (compared the value read off by experts), the average error is about 0.5. These results indicate a practicable value.

• 한국도로학회논문집
• /
• 제15권5호
• /
• pp.177-185
• /
• 2013

PURPOSES : This study attempts to design and establish the road surface condition detection system by using the image processing that is expected to help implement the low-cost and high-efficiency road information detection system by examining technology trends in the field of road surface condition information detection and related case studies. METHODS : Adapted visual information collecting method(setting a stereo camera outside of the vehicle) and visual information algorithm(transform a Wavelet Transform, using the K-means clustering) Experiments and Analysis on Real-road, just as four states(Dry, Wet, Snow, Ice). RESULTS : Test results showed that detection rate of 95% or more was found under the wet road surface, and the detection rate of 85% or more in snowy road surface. However, the low detection rate of 30% was found under the icy road surface. CONCLUSIONS : As a method to improve the detection rate of the mobile road surface condition information detection system developed in this study, more accurate phase analysis in the image processing process was needed. If periodic synchronization through automatic settings of the camera according to weather or ambient light was not made at the time of image acquisition, a significant change in the values of polarization coefficients occurs.

• 한국멀티미디어학회논문지
• /
• 제22권3호
• /
• pp.349-356
• /
• 2019

When a prescription change occurs in the hospital depending on a patient's improvement status, pharmacists directly classify manually returned pills which are not taken by a patient. There are hundreds of kinds of pills to classify. Because it is manual, mistakes can occur and which can lead to medical accidents. In this study, we have compared YOLO, Faster R-CNN and RetinaNet to classify and detect pills. The data consisted of 10 classes and used 100 images per class. To evaluate the performance of each model, we used cross-validation. As a result, the YOLO Model had sensitivity of 91.05%, FPs/image of 0.0507. The Faster R-CNN's sensitivity was 99.6% and FPs/image was 0.0089. The RetinaNet showed sensitivity of 98.31% and FPs/image of 0.0119. Faster RCNN showed the best performance among these three models tested. Thus, the most appropriate model for classifying pills among the three models is the Faster R-CNN with the most accurate detection and classification results and a low FP/image.

• 치과방사선
• /
• 제22권1호
• /
• pp.128-136
• /
• 1992

This study was undertakened to investigate the change of image characteristics on dental films according to the process of using time of processing solution in automatic processor. Base + fog density, film density and subject contrast were measured with the digital densitometer, the pH of developing and fixing solution were measured with Digital pH / ION Meter. The following results were obtained: 1. Base + fog density was increased with the process of using time of the processing solution and was over the maximum permissible base + fog density 0.25 from the 3rd day. 2. Film density was increased with the process of using time of the processing solution. 3. Subject contrast was decreased with the process of using time of the processing solution. 4. The pH of the developing solution was decreased with the process of using time, the pH of the fixing solution was increased.

• 한국측량학회지
• /
• 제36권6호
• /
• pp.525-534
• /
• 2018

카메라 캘리브레이션은 카메라의 초점거리, 주점위치, 렌즈왜곡 등의 매개변수를 결정하는 작업으로 이를 위해서 주로 체커보드를 촬영한 영상을 사용하고 있다. 체커보드 영상에서 타겟을 자동으로 인식할 때 기존의 연구는 사용자가 타겟인식을 위한 입력 매개변수를 잘 이해하고 있어야 하거나 영상에서 체커보드가 모두 나타나야 하는 한계점이 있었다. 이에 본 연구에서는 체커보드 중심부와 외곽부분에 각각 4개씩 8개의 블랍을 포함하는 직사각형을 이용하여 체커보드 영상의 일부만 촬영된 경우에도 자동으로 타겟점의 번호를 부여할 수 있고 별도의 입력 매개 변수 없이 자동으로 타겟을 인식하는 방법을 제안하였다. 본 연구에서 체커보드 타겟의 중심점을 자동으로 추출하기 위해서 흑백패턴의 왜곡, 경계선 변화빈도, 흑백픽셀의 비율의 3가지 조건을 이용하였다. 또한 체커보드의 방향성과 번호부여는 블랍을 이용하였다. 두 가지 타입의 체커보드에 대한 실험을 통해서 36장의 영상에 대해 1분 이내의 짧은 시간에 체커보드 타겟을 자동으로 인식할 수 있었다.

• 한국작물학회:학술대회논문집
• /
• 한국작물학회 2022년도 추계학술대회
• /
• pp.80-80
• /
• 2022

Estimation of heading date of paddy rice is laborious and time consuming. Therefore, automatic estimation of heading date of paddy rice is highly essential. In this experiment, deep learning classification models were used to classify two difference categories of rice (vegetative and reproductive stage) based on the panicle initiation of paddy field. Specifically, the dataset includes 444 slanted view images belonging to two categories and was then expanded to include 1,497 images via IMGAUG data augmentation technique. We adopt two transfer learning strategies: (First, used transferring model weights already trained on ImageNet to six classification network models: VGGNet, ResNet, DenseNet, InceptionV3, Xception and MobileNet, Second, fine-tuned some layers of the network according to our dataset). After training the CNN model, we used several evaluation metrics commonly used for classification tasks, including Accuracy, Precision, Recall, and F1-score. In addition, GradCAM was used to generate visual explanations for each image patch. Experimental results showed that the InceptionV3 is the best performing model in terms of the accuracy, average recall, precision, and F1-score. The fine-tuned InceptionV3 model achieved an overall classification accuracy of 0.95 with a high F1-score of 0.95. Our CNN model also represented the change of rice heading date under different date of transplanting. This study demonstrated that image based deep learning model can reliably be used as an automatic monitoring system to detect the heading date of rice crops using CCTV camera.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2002년도 춘계학술대회 논문집
• /
• pp.369-372
• /
• 2002

In this research, a vision system for detecting tool breakage which is hardly detected by such indirect in-process measurement method as acoustic emission, cutting torque and motor current was developed and embedded into a PC-NC system. The vision system consists of CMOS image sensors, a slit beam laser generator and an image grabber board. Slit beam laser was emitted on the tool surface to separate the tool geometry well from the various obstacles surrounding the tool. An image of tool is captured through two steps of signal processing, that is, median filtering and thresholding and then the tool is estimated normal or broken by use of change of the centroid of the captured image. An air curtain made by the jetting high-pressure air in front of the lens was devised to prevent the vision system from being contaminated by scattered coolant, cutting chips in cutting process. To embed the vision system to a Siemens PC-NC controller 840D NC, an HMI(Human Machine Interface) program was developed under the Windows 95 operating system of MMC103. The developed HMI is placed in a sub window of the main window of 840D and this program can be activated or deactivated either by a soft key on the operating panel or M codes in the NC part program. As the tool breakage is detected, the HMI program emit a command for automatic tool change or send alarm to the NC kernel. Evaluation test in a high speed tapping center showed the developed system was successful in detection of the small-radius tool breakage.

• 한국정보통신학회논문지
• /
• 제21권1호
• /
• pp.107-114
• /
• 2017

홍채진단은 홍채의 패턴, 색 등 다른 특징들을 조사하여 환자의 병을 진단하는 대체의학이다. 이 논문에서는 촬영한 홍채이미지의 차영상을 이용해 홍채를 분석하고 홍채 변화에 따른 환자의 건강진단에 활용한 질병예측 알고리즘을 제안한다. 그러나 기존의 연구는 홍채영상을 이용하여 홍채 내의 특정 패턴을 검출하는 알고리즘 연구로 홍채의 다양한 정보로부터 건강 상태를 체크하는 진단시스템으로 사용하기에는 부족하다. 따라서 이 논문에서는 촬영된 홍채영상의 차영상을 이용해 질병의 조기 진단 및 질병의 전개과정을 명확히 판단한다. 또한 홍채영상으로부터 8가지 주요 홍채병소징후를 추출하고 검진의 정확도를 실험한 결과 패턴 매칭 기법에 의한 인식률 91%로 홍채진단의 자동화에 적용 가능하다.