• Journal of Software Engineering Society
• /
• v.29 no.1
• /
• pp.13-21
• /
• 2020

As CNN model is applied to various domains such as image classification and object detection, the performance of CNN model which is used to safety critical system like autonomous vehicles should be reliable. To evaluate that CNN model can sustain the performance in various environments, we developed an image data augmentation apparatus which generates images that is changed background. If an image which contains object is entered into the apparatus, it extracts an object image from the entered image and generate s composed images by synthesizing the object image with collected background images. A s a method to evaluate a CNN model, the apparatus generate s new test images from original test images, and we evaluate the CNN model by the new test image. As a case study, we generated new test images from Pascal VOC2007 and evaluated a YOLOv3 model with the new images. As a result, it was detected that mAP of new test images is almost 0.11 lower than mAP of the original test images.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.3
• /
• pp.693-698
• /
• 2018

This paper proposes a vehicle detection method for parking areas using unmanned aerial vehicles (UAVs) and using YOLOv2, which is a recent, known, fast, object-detection real-time algorithm. The YOLOv2 convolutional network algorithm can calculate the probability of each class in an entire image with a one-pass evaluation, and can also predict the location of bounding boxes. It has the advantage of very fast, easy, and optimized-at-detection performance, because the object detection process has a single network. The sliding windows methods and region-based convolutional neural network series detection algorithms use a lot of region proposals and take too much calculation time for each class. So these algorithms have a disadvantage in real-time applications. This research uses the YOLOv2 algorithm to overcome the disadvantage that previous algorithms have in real-time processing problems. Using Darknet, OpenCV, and the Compute Unified Device Architecture as open sources for object detection. a deep learning server is used for the learning and detecting process with each car. In the experiment results, the algorithm could detect cars in a dense area using UAVs, and reduced overhead for object detection. It could be applied in real time.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.5 s.37
• /
• pp.87-94
• /
• 2005

This paper describes the method of detecting the region of vehicle number plate in colored car image with number plate. Vehicle number plate region generally shows formula colors in accordance with type of car. According to this, we use the method to combine a color ingredient H of HSI color model and a color ingredient Q of YIQ color model. However, the defect which a total operation time takes much exists if it uses such method. Therefore, in this paper, the concurrent accomplishes a candidate area extraction operation as draw a color H and Q ingredient among steps of extracting a region of vehicle number Plate. After the above step, as a next step in combination with color H and Q we can accomplish an region extraction fast by comparing to candidate regions extracted from each steps not to do a comparison operation to all of image pixel information. We also show implementation results Processed at each steps and compare with extraction time according to image resolutions.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.8-10
• /
• 2019

열화상 장비는 빛이 없는 암흑 상태에서도 물체에서 발산하는 적외선을 탐지하여 이를 영상으로 제공하는 장비이다. 이러한 장점으로 기존 활용되던 군사 분야와 더불어 자동차 및 감시시스템 등 다양한 민수 분야로 활용분야가 넓어지고 있다. 따라서 기존 방식인 FPGA 기반 열화상 이미지 모듈은 민수 시장의 다양한 요구사항과 환경을 반영하기에는 힘들 실정이다. 그에 따라 FPGA 기반 시스템의 단점을 보완하고 추가적인 요구사항을 만족하는 시스템의 필요성이 대두되었다. 본 논문에서는 상용 이미지 프로세서 기반 열화상 이미지 처리 모듈을 제안한다. 기존 FPGA 기반 열화상 이미지 처리 방식이 아닌 상용 이미지 프로세서 기반 구조 설계로 함으로써 다양한 영상 입·출력 인터페이스 수신 및 표준 영상 출력 포멧을 지원한다. 따라서 상용 프로세서 기반 열상 처리 모듈을 통한 시스템 개발 시 뛰어난 접근성으로 시스템 구축이 용이하고 다양한 요구사항 적용이 가능함에 따라 개발 기간 및 비용 단축, 다양한 응용에 사용이 가능할 것으로 예상한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2002.11a
• /
• pp.81-84
• /
• 2002

멀티미디어 분야가 급성장하면서 좀더 효율적으로 멀티미디어 자료의 저장, 처리, 검색을 위한 연구가 진행되고 있다. 특히, 내용기반 시각정보 검색에 있어 지능형 시스템(Intelligent System)을 접목하여 의미적 접근을 시도하는 I-CBIR(Intelligent-Content Based Image Retrieval)에 관한 연구가 진행되고 있다. 또한, 내용기반 이미지검색 시스템에 온톨로지(Ontology)의 이론을 적용하여 이미지에 의미를 부여하여 개념적 검색이 가능하도록 노력하고 있다. 이러한 연구에서 적용된 대형의 온톨로지는 이미지 검색 시스템에 적합하지 않게 너무 방대한 정보를 가지고 있으며, 또한 시대적 변화에 대응하지 못하여 I-CBIR 시스템에서 그 효율성을 제대로 발휘하지 못하고 있다. 따라서 본 논문에서는 많은 대형 온톨로지 중에서 WordNet을 선택하여, WordNet의 구축 방법에 기반한 자동차(Car)에 대한 도메인 온톨로지(Domain Ontology)를 구축해보고, 구축된 도메인 온톨로지를 적용함으로써 더 향상된 I-CBIR 시스템이 되도록 하였다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.01a
• /
• pp.69-70
• /
• 2021

본 논문은 PPM으로 관리되고 있는 자동차 부품 제조 공정에서 검사자의 육안검사 방법을 대체하기 위해 머신비전 및 CNN 기반 불량 검출 시스템으로 제안되었던 방식들의 단점을 개선하기 위하여 기존 머신 비전 기술에 합성곱 오토인코더 모델을 적용하여 단점을 해결하였다. 본 논문에서 제시한 오토인코더를 이용하는 방법은 정상 생산품의 이미지만으로 학습을 진행하고, 학습된 모델은 불량 부위가 포함된 이미지를 입력받아 정상 이미지로 출력한다. 이 방법을 사용하여 불량의 부위와 크기를 알 수 있었으며 불량 여부의 판단은 임계치에 의한 불량 부위의 화소 수 계산으로 판단하였다.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.15 no.4
• /
• pp.27-32
• /
• 2007

Head-Up Display system makes it possible for the driver to be informed of important vehicle data such as vehicle speed, engine RPM or navigation data without taking the driver's eyes off the road. Long focal length optics, LCD with bright illumination, image generator and vehicle interface controllers are key parts of head-up display system. All these parts have been designed, developed and applied to the test vehicle. Virtual images are located about 2m ahead of the driver's eye by projecting it onto the windshield just below the driver's line of sight. Developed head-up display system shows satisfactory results for future commercialization.

• Journal of Distribution Science
• /
• v.8 no.1
• /
• pp.43-52
• /
• 2010

Country images of Korea and Japan based on economic development, education level, goods' quality, R&D, political democratization and quality of life, perceived by Chinese university students in Beijing, Shanghai and Shandong province of chinese emerging markets as the representative of a potential buying power group, are surveyed, which, then are used to study how the perceived country images effect on their purchasing intention for Korean and Japanese products, such as, foods, cars, fashions, music CDs, electronic products and living goods. The study shows that, in chinese emerging markets, country image affects on the purchase intention of each products differently. The country image of Korea was less influential than that of Japan on the Chines students' purchasing intention for the goods other than the electronic goods. Despite the small number of the sample, this study showed the importance of country image in the in chinese emerging markets and suggested the need for both the government and private sector to take a strategy to enhance the country image by finding the relation between the elements of country image and the intention to purchase certain product.

• Journal of the Institute of Convergence Signal Processing
• /
• v.12 no.4
• /
• pp.303-308
• /
• 2011

Vehicle black boxes that have similar functions as airplane black boxes are currently being used due to the loss of many lives and properties arising from vehicle accidents. Both black-box products and Event Data Recorder(EDR) systems are currently available in the market. Most of the existing in-vehicle black boxes, however, record only external videos and images and cannot show the vehicle's driving status, whereas EDR products record only the driving status and not external videos. To address the problem of black boxes that can record only videos and images and that of EDR systems that can record only driving data, an integrated vehicle state and video recording system that uses MOST(Media-oriented System Transport) and OBD-II(Onboard Diagnostics II) and CAM (camera) and GPS (global positioning system).

• Journal of Internet of Things and Convergence
• /
• v.9 no.1
• /
• pp.65-70
• /
• 2023

In this paper, we propose an autonomous driving system using an end-to-end model to improve lane departure and misrecognition of traffic lights in a vision sensor-based system. End-to-end learning can be extended to a variety of environmental conditions. Driving data is collected using a model car based on a vision sensor. Using the collected data, it is composed of existing data and data with outlayers removed. A class was formed with camera image data as input data and speed and steering data as output data, and data learning was performed using an end-to-end model. The reliability of the trained model was verified. Apply the learned end-to-end model to the model car to predict the steering angle with image data. As a result of the learning of the model car, it can be seen that the model with the outlayer removed is improved than the existing model.