• 제어로봇시스템학회논문지
• /
• 제15권9호
• /
• pp.922-928
• /
• 2009

This thesis puts forward a suggestion of measuring inner crack and center segregation in steel processing by using scanner and image processing with sulfur printer. To sum up, there are three points in this system. First, it scans sulfur printer and process the image by using histogram, image processing, and the mask. Second, it measures crack and center segregation by the fictitious image and output the length, thickness, exponent and grade on the mornitor. And finally, it gathers the measurement result image and data at the server and this information is used as data for the next casting.

• 한국군사과학기술학회지
• /
• 제9권2호
• /
• pp.111-117
• /
• 2006

This paper describes the design principles and methods of electronic processor for thermal imager with 320$\times$240 staring array infrared detector. For the detector's nonuniformity correction and excellent image quality, we have designed the multi-point correction method using the defocusing technique of the optics. And to enhance the image of low contrast and improve the detection capability, the new technique of histogram processing has been designed. Through these image processing techniques, we have developed the high quality thermal imager and acquired a satisfactory thermal image. The result of MRTD(Minimum Resolvable Temperature Difference) is $0.1^{\circ}C$ at 4cycles/mard.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2002년도 Proceedings of International Symposium on Remote Sensing
• /
• pp.45-48
• /
• 2002

Satellite image processing software should be able to support interoperability that is to enable consumers, citizens, large and small businesses and governments to make the fullest possible use of the world's spatial data and spatial processing resources. The goal of interoperability is to perform hands-on research, development, testing, fielding, and validation of potential and existing interoperable geoprocessing technologies. In this paper, component based architecture for fast interoperability is designed to maximize interoperability among the OpenGIS components specified in the OpenGIS abstract specifications and implementation specifications and to mix up the OpenGIS components and the fast access mechanism for high performance.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2019년도 춘계학술발표대회
• /
• pp.401-403
• /
• 2019

In this paper, we propose a generative adversarial networks (GAN) based text-to-image generating method. In many natural language processing tasks, which word expressions are determined by their term frequency -inverse document frequency scores. Word2Vec is a type of neural network model that, in the case of an unlabeled corpus, produces a vector that expresses semantics for words in the corpus and an image is generated by GAN training according to the obtained vector. Thanks to the understanding of the word we can generate higher and more realistic images. Our GAN structure is based on deep convolution neural networks and pixel recurrent neural networks. Comparing the generated image with the real image, we get about 88% similarity on the Oxford-102 flowers dataset.

• Journal of Information Processing Systems
• /
• 제16권6호
• /
• pp.1479-1494
• /
• 2020

As an important part of image processing, image demosaicking has been widely researched. It is especially necessary to propose an efficient interpolation algorithm with good visual quality and performance. To improve the limitations of residual interpolation (RI), based on RI algorithm, minimalized-Laplacian RI (MLRI), and iterative RI (IRI), this paper focuses on adaptive RI (ARI) and proposes an improved ARI (IARI) algorithm which obtains more distinct R, G, and B colors in the images. The proposed scheme fully considers the brightness information and edge information of the image. Since the ARI algorithm is not completely adaptive, IARI algorithm executes ARI algorithm twice on R and B components according to the directional difference, which surely achieves an adaptive algorithm for all color components. Experimental results show that the improved method has better performance than other four existing methods both in subjective assessment and objective assessment, especially in the complex edge area and color brightness recovery.

• Journal of Multimedia Information System
• /
• 제9권3호
• /
• pp.183-190
• /
• 2022

Image-based 3D object detection is one of the important and difficult problems in autonomous driving and robotics, and aims to find and represent the location, dimension and orientation of the object of interest. It generates three dimensional (3D) bounding boxes with only 2D images obtained from cameras, so there is no need for devices that provide accurate depth information such as LiDAR or Radar. Image-based methods can be divided into three main categories: monocular, stereo, and multi-view 3D object detection. In this paper, we investigate the recent state-of-the-art models of the above three categories. In the multi-view 3D object detection, which appeared together with the release of the new benchmark datasets, NuScenes and Waymo, we discuss the differences from the existing monocular and stereo methods. Also, we analyze their performance and discuss the advantages and disadvantages of them. Finally, we conclude the remaining challenges and a future direction in this field.

• 국제학술발표논문집
• /
• The 3th International Conference on Construction Engineering and Project Management
• /
• pp.224-229
• /
• 2009

Digital image recognition has been used for steel bridge surface assessment since late 1990s. However, the non-uniform illumination problems such as shades, shadows, and highlights are still challenges in image processing to date. Therefore, this paper develops a new approach to tackle the non-uniform illumination problem for rust image adjustment. The inhomogeneous illumination problem is divided into shades/shadows and highlights in this paper. The proposed BEMD-morphology approach (BMA) utilizes the bidimensional empirical mode decomposition to mitigate the shade/shadow effect, and the morphological processing to detect and replace the highlight area. Finally, the rust image processed with the BMA will be segmented by the K-Means algorithm, one of the most popular and effective methods, to show the effectiveness of illumination adjustment.

• 지적과 국토정보
• /
• 제50권2호
• /
• pp.155-168
• /
• 2020

오픈소스는 4차 산업혁명의 핵심 성장 동력으로서 다양한 영상해석 알고리즘의 지속적인 개발과 활용이 기대되고 있다. 본 연구의 목적은 UAS 영상해석 오픈소스 기반 알고리즘의 3차원 재현 중 물의 재현 및 이동체 필터링 기능과 데이터 처리 소요시간을 중점으로 비교·분석하여 효용성을 검토하는 것이다. 5가지 매칭 알고리즘을 'ANN-Benchmarks' 프로그램을 통해 재현율 및 처리속도 기준으로 비교하였고 HNSW(hierarchical navigable small world) 매칭 알고리즘이 가장 양호한 것으로 판단하였다. 이를 바탕으로 삼각측량, 점군 데이터 조밀화, 표면생성의 단계별 기법들을 조합하여 108가지 영상해석 알고리즘을 구성하였다. 또한, 바다와 인접한 공원의 UAS(unmanned aerial system) 영상을 대상으로 108가지 영상해석 알고리즘의 3차원 재현 및 데이터 처리 소요시간을 고찰하고 상업용 영상해석 소프트웨어 'Pix4D Mapper'와 비교·분석하였다. 연구 결과, 3차원 재현 중 물의 재현 및 이동체 필터링 기능 면에서 양호한 알고리즘을 각각 특정하였고 소요시간이 가장 낮은 알고리즘을 선정, 'Pix4D Mapper' 처리 결과와 비교하여 알고리즘의 효용성을 입증하였다.

• Journal of Biosystems Engineering
• /
• 제37권2호
• /
• pp.122-129
• /
• 2012

Purpose: Recent mechatronics technology is the most appropriate high technology in agricultural applications to save repetitious labor. Method: Cow's body parameters were measured by several traditional measurers. Image processing technology was used to measure automatically their parameters to reduce labor and time. The parameters were measured form a small model cow which is easily measured, instead to a real cow. The image processing system designed and built for this project was composed of a PC, grabber card, and two cameras, which are located on the side and the top of the model cow. Tests of verification had measured 10 dairy cows. Result: Nine parameters of the model cow's body were measured, and the difference between the real data and the data by image processing was less than 16.7%. Based on the results of the research, the parameters of a real cow had measured of chest depth, withers height, Pelvic arch height, body length, slope body length, chest width, hip width, thurl width, and pin bone width were compared with image processing data. Conclusions: In the Demonstration test, Result had obtained similar data of cow model experiments, and the most of errors were shown less than 5% relatively good result.

• 한국구조물진단유지관리공학회 논문집
• /
• 제20권4호
• /
• pp.104-110
• /
• 2016

본 연구에서는 매립관 및 공동에 대한 GPR 탐사의 실증적 자료 확보 및 GPR 탐사 자료의 이미지 처리를 위한 프로그램을 개발하기 위한 토조의 GPR 탐사를 시행하였다. 비포장 및 아스팔트 포장의 경우 심도 0.7m에서 물채움 공동을 확인할 수 있었으나, 1.0 m 및 1.3 m 깊이에 매립한 공동은 탐지 불가하였다. 무근 및 철근 콘크리트 포장의 경우 신호의 간섭이 심하여 공동을 탐지하는 것이 어려웠다. GPR 탐사 데이터의 이미지 처리를 위한 프로그램 GPRiPP을 개발하였다. 주요한 처리 기법인 백그라운드 제거, 스택킹 및 게인 기능 알고리즘을 프로그램에 반영하였으며, 적절한 백그라운드 제거 및 게인 기능을 사용하면 기존 프로그램과 유사한 이미지 처리 결과를 도출할 수 있음을 확인하였다.