• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.463-465
• /
• 2021

Currently, the inspection method printed on food packages and boxes is to sample only a few products and inspect them with human eyes. Such a sampling inspection has the limitation that only a small number of products can be inspected. Therefore, accurate inspection using a camera is required. This paper proposes a deep learning object recognition technology model, which is an artificial intelligence technology, as a method for detecting the defects of expiration date printed on the product packaging. Using the Faster R-CNN (region convolution neural network) model, the color images, converted gray images, and converted binary images of the printed expiration date are trained and then tested, and each detection rates are compared. The detection performance of expiration date printed on the package by the proposed method showed the same detection performance as that of conventional vision-based inspection system.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.33 no.1
• /
• pp.40-45
• /
• 2013

The purpose of this study was to detect the pin hole defect of dental composite restoration using lock-in thermography method. Amplitude and phase images of the composite resin specimens were analyzed according to the lock-in frequency and the diameter of defect area. Through the amplitude image analysis, at lock-in frequency of 0.05 Hz, defect diameters 2-5 mm exhibited the highest amplitude contrast value between defective area and sound area. The lock-in frequency range of 0.3-0.5 Hz provided good phase angle contrast for the defect area. At lock-in frequency range of 0.5 Hz, defect diameter of 5 mm exhibited the highest phase contrast value. It is concluded that the infrared lock-in thermography method verified the effectiveness for detecting the pin hole defect of dental composite restoration.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.05a
• /
• pp.446-450
• /
• 2020

딥러닝 기반의 영상 분석 방법들은 많은 양의 학습 데이터가 필요하며, 학습 데이터 구축에는 많은 시간과 노력이 소요된다. 특히 객체 검출 분야의 경우 영상 내 객체의 위치, 크기, 범주 등의 정보가 모두 필요하여 학습 데이터 구축에 더 많은 어려움이 있으며, 이를 해결하기 위해 최근 이미지 합성기반 데이터 증강에 관한 연구가 활발히 진행되고 있다. 이미지 합성기반 데이터 증강 방법은 배경 영상에 객체를 합성할 때 객체와 배경 영상이 접한 영역에서 아티팩트(Artifact)가 발생하며, 이는 객체 검출 모델이 아티팩트를 객체의 특징으로 모델링하여 검출 성능이 저하되는 원인이 된다. 이러한 문제를 해결하기 위하여 본 논문에서는 양방향 필터 기반의 이미지 합성 방법을 제안하고, 단일 단계 검출의 대표적인 방법인 RetinaNet을 이용하여 이미지 합성기반 데이터 증강 방법의 성능을 분석하였다. 공개 데이터셋에 대한 실험 결과 본 논문에서 사용한 단일 검출 방법 및 데이터 증강 기법을 사용하면 더 적은 양의 증강 데이터로 기존 방법과 동일한 성능을 보여주는 것을 확인하였다.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.5
• /
• pp.211-220
• /
• 2022

Recently, research on smart factories triggered by the 4th industrial revolution is being actively conducted. Accordingly, the manufacturing industry is conducting various studies to improve productivity and quality based on deep learning technology with robust performance. This paper is a study on the method of detecting tire surface defects in the visual inspection stage of the tire manufacturing process, and introduces a tire surface defect detection method using a depth image acquired through a 3D camera. The tire surface depth image dealt with in this study has the problem of low contrast caused by the shallow depth of the tire surface and the difference in the reference depth value due to the data acquisition environment. And due to the nature of the manufacturing industry, algorithms with performance that can be processed in real time along with detection performance is required. Therefore, in this paper, we studied a method to normalize the depth image through relatively simple methods so that the tire surface defect detection algorithm does not consist of a complex algorithm pipeline. and conducted a comparative experiment between the general normalization method and the normalization method suggested in this paper using YOLO V3, which could satisfy both detection performance and speed. As a result of the experiment, it is confirmed that the normalization method proposed in this paper improved performance by about 7% based on mAP 0.5, and the method proposed in this paper is effective.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.9 no.3
• /
• pp.145-153
• /
• 2009

The purpose of this research is to propose a system to detect a strip defect on a iron plate using an image processing, one way of finding defects on an iron plate. An existing way of image processing is using a light source which release a light energy in a certain frequency and a light absorbing display which responds to the light source. This research attempts to detect defects by using a image processing and multi-Tread which handles an illumination, without depending on characteristics of light frequency. One of the advantages of this method is that it makes up for the weakness of the existing method which was too difficult for users to notice a defect. Also this method makes it possible to realize a real-time monitoring on a plate of iron. The other advantage of this method is that it reduces the price of hardwares on demand to match the frequency of light emitting display and light absorbing display because this method only needs a hardware which is easy to buy in any market.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.47 no.2
• /
• pp.11-20
• /
• 2010

In this paper we propose a face detection system which consists of a method of face candidate extraction using skin color and a method of face verification using the feature of facial structure. Firstly, the proposed extraction method of face candidate uses the image segmentation and merging algorithm in the regions of skin color and the neighboring regions of skin color. These two algorithms make it possible to select the face candidates from the variety of faces in the image with complicated backgrounds. Secondly, by using the partial face classifier, the proposed face validation method verifies the feature of face structure and then classifies face and non-face. This classifier uses face images only in the learning process and does not consider non-face images in order to use less number of training images. In the experimental, the proposed method of face candidate extraction can find more 9.55% faces on average as face candidates than other methods. Also in the experiment of face and non-face classification, the proposed face validation method obtains the face classification rate on the average 4.97% higher than other face/non-face classifiers when the non-face classification rate is about 99%.

• Journal of IKEEE
• /
• v.22 no.4
• /
• pp.1079-1087
• /
• 2018

Recently, computer vision field based deep learning artificial intelligence has become a hot topic among various image analysis boundaries. In this study, flames are detected in fire images using the Faster R-CNN algorithm, which is used to detect objects within the image, among various image recognition algorithms based on deep learning. In order to improve fire detection accuracy through a small amount of data sets in the learning process, we use image augmentation techniques, and learn image augmentation by dividing into 6 types and compare accuracy, precision and detection rate. As a result, the detection rate increases as the type of image augmentation increases. However, as with the general accuracy and detection rate of other object detection models, the false detection rate is also increased from 10% to 30%.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.3
• /
• pp.171-179
• /
• 2020

Recently, studies have shown that convolution neural networks are achieving the best performance in image classification, object detection, and image generation. Vision based defect inspection which is more economical than other defect inspection, is a very important for a factory automation. Although supervised anomaly detection algorithm has far exceeded the performance of traditional machine learning based method, it is inefficient for real industrial field due to its tedious annotation work, In this paper, we propose ADGAN, a unsupervised anomaly detection architecture using the variational autoencoder and the generative adversarial network which give great results in image generation task, and demonstrate whether the proposed network architecture identifies anomalous images well on MNIST benchmark dataset as well as our own welding defect dataset.

• Journal of the Korea Convergence Society
• /
• v.8 no.8
• /
• pp.33-40
• /
• 2017

In this paper, we propose a text extraction method by pattern analysis on contour for effective text detection in image. Text extraction algorithms using edge based methods show good performance in images with simple backgrounds, The images of complex background has a poor performance shortcomings. The proposed method simplifies the color of the image by using K-means clustering in the preprocessing process to detect the character region in the image. Enhance the boundaries of the object through the High pass filter to improve the inaccuracy of the boundary of the object in the color simplification process. Then, by using the difference between the expansion and erosion of the morphology technique, the edges of the object is detected, and the character candidate region is discriminated by analyzing the pattern of the contour portion of the acquired region to remove the unnecessary region (picture, background). As a final result, we have shown that the characters included in the candidate character region are extracted by removing unnecessary regions.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.35 no.3
• /
• pp.193-199
• /
• 2015

Active thermography techniques have the capability of inspecting a broad range simultaneously. By evaluating the phase difference between the defected area and the healthy area, the technique indicates the qualitative location and size of the defect. Previously, the development of the defect detection method used a variety of materials and the test specimen was done. In this study, the proposed technique of lock-in is verified with artificial specimens that have different size and depth of subsurface defects. Finally, the defect detection capability was evaluated using comparisons of the phase image and the amplitude image according to the size and depth of defects.