• 반도체디스플레이기술학회지
• /
• 제19권2호
• /
• pp.11-15
• /
• 2020

In a typical initial setup of a PCB component inspection system, operators should manually input various information such as category, position, and inspection area for each component to be inspected, thus causing much inconvenience and longer setup time. Although there are many deep learning based object detectors, RetinaNet is regarded as one of best object detectors currently available. In this paper, a method using an extended RetinaNet is proposed that automatically detects its component category and position for each component mounted on PCBs from a high-resolution color input image. We extended the basic RetinaNet feature pyramid network by adding a feature pyramid layer having higher spatial resolution to the basic feature pyramid. It was demonstrated by experiments that the extended RetinaNet can detect successfully very small components that could be missed by the basic RetinaNet. Using the proposed method could enable automatic generation of inspection areas, thus considerably reducing the setup time of PCB component inspection systems.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권11호
• /
• pp.4557-4572
• /
• 2020

The Alaska competition provides an opportunity to study the practical problems of real-world steganalysis. Participants are required to solve steganalysis involving various embedding schemes, inconsistency JPEG Quality Factor and various processing pipelines. In this paper, we propose a method to ensemble multiple deep learning steganalyzers. We select SRNet and RESDET as our base models. Then we design a three-layers model ensemble network to fuse these base models and output the final prediction. By separating the three colors channels for base model training and feature replacement strategy instead of simply merging features, the performance of the model ensemble is greatly improved. The proposed method won second place in the Alaska 1 competition in the end.

• 한국조경학회지
• /
• 제52권2호
• /
• pp.39-50
• /
• 2024

색채는 중요한 시각적 요소로서 도시 이미지와 사람들의 인식 형성에 중요한 영향을 미친다. 도시환경에서 색채를 정량적으로 분석하는 작업은 복잡한 과정을 필요로 하여 과거에는 실행하기가 어려웠다. 그러나 최근 머신 러닝 기술의 급속한 발전으로 관광객이 공유한 사진을 이용하여 도시 색채를 분석하는 것이 가능해졌다. 본 연구는 중국의 인기 관광지인 대리시를 사례로 선정하여 관광객이 공유한 대리시의 사진을 수집하였으며, 머신 러닝 기술을 결합하여 대규모 도시 색채를 측정하는 방법을 탐색하였다. 구체적으로는 먼저 DeepLabv3+ 모델을 사용하여 ADE20k 데이터 셋을 기반으로 관광객이 공유한 사진의 의미 분할을 수행하여 사진에서 인공 요소를 분리했다. 다음으로 K-means 클러스터링 알고리즘을 사용하여 대리시의 인공 요소의 주요 색상을 추출하고, 이러한 색상 간의 상관관계를 분석하기 위해 인접 매트릭스를 구축했다. 연구 결과에 따르면 대리시의 인공 요소의 주요 색상은 주황-회색이 가장 높은 비율을 차지한다. 또한, 회색 계열의 색상이 다른 색상과 자주 조합되어 사용되는 경향이 있다. 분석에 따르면 대리시의 인공 요소의 색채 특성은 지역의 민족 문화와 불교 문화의 영향을 받는 것으로 나타났다. 본 연구는 색채 분석을 위한 새로운 접근 방법을 제공하며, 연구 결과는 대리시가 관광객의 기대에 부합하는 도시 색채 이미지를 형성하는 데 도움이 될 뿐만 아니라 향후 대리시의 색채 계획을 위한 참고 자료를 제공하고자 한다.

• 한국환경복원기술학회지
• /
• 제26권6호
• /
• pp.85-96
• /
• 2023

This study proposes a method for forest vegetation monitoring using high-resolution aerial imagery captured by unmanned aerial vehicles(UAV) and deep learning technology. The research site was selected in the forested area of Mountain Dogo, Asan City, Chungcheongnam-do, and the target species for monitoring included Pinus densiflora, Quercus mongolica, and Quercus acutissima. To classify vegetation species at the pixel level in UAV imagery based on characteristics such as leaf shape, size, and color, the study employed the semantic segmentation method using the prominent U-net deep learning model. The research results indicated that it was possible to visually distinguish Pinus densiflora Siebold & Zucc, Quercus mongolica Fisch. ex Ledeb, and Quercus acutissima Carruth in 135 aerial images captured by UAV. Out of these, 104 images were used as training data for the deep learning model, while 31 images were used for inference. The optimization of the deep learning model resulted in an overall average pixel accuracy of 92.60, with mIoU at 0.80 and FIoU at 0.82, demonstrating the successful construction of a reliable deep learning model. This study is significant as a pilot case for the application of UAV and deep learning to monitor and manage representative species among climate-vulnerable vegetation, including Pinus densiflora, Quercus mongolica, and Quercus acutissima. It is expected that in the future, UAV and deep learning models can be applied to a variety of vegetation species to better address forest management.

• 한국염색가공학회:학술대회논문집
• /
• 한국염색가공학회 2006년도 춘계학술발표회 논문집
• /
• pp.210-214
• /
• 2006

• 복식
• /
• 제56권9호
• /
• pp.98-111
• /
• 2006

The objective of this research is to examine the commonness and differences of Korean, Chinese and Japanese costume colors of modern and present ages. The result of this study showed that modern China and Japan had quick influx speed of Western culture. Dissimilarly, modern Korea kept conception of colors from Chosun periods that show the high frequency of 'Five Elements Colors' and neutral colors in Red, Yellow and Purple Blue. Today, the costumes of China, Korea and Japan use similar tones of color but each country approached in different selections of achromatic colors; Korean prefers color in Yellow Red, Purple, and Chinese in Green Yellow, Green and Japanese in Purple Blue. Light greyish and pale toned Yellow Red and grayish tone have increased in modern Chinese and Japanese costumes. Also both countries have corresponding assumptions in using color of Red in strong tone. The analysis of color and tone distribution showed that, Japanese costume colors in modern and present times have correlative number of use as in Western culture. Traditionally, Japan has least notion of using 'Five Elements Colors' which only gives minor changes by convergence of Western color culture. In other side, China had developed in color rather than tone compares to Korea and Japan by using many of the Red color of strong, vivid and deep tones which made red distinguishing color of China. Japan continues to use of low chroma colors and became a characteristic in modern and present day, also they use an abundance of color in Yellow Red, purple Blue. Korea has a higher frequency showing in light, bright tones of color distinctively compares to China and Japan.

• 한국염색가공학회지
• /
• 제11권2호
• /
• pp.38-45
• /
• 1999

Natural red colorants were extracted from Phytolacca americana Linne by using 50% ethanol solution at room temperature for 12 hours. The colorant components were partially purified as yellow and deep red colorants by thin layer chromatography. Natural red colorants were consisted of major water-soluble red colorant, having maximum absorbance at 538nm and alcohol-soluble yellow colorant, having maximum absorbance at 664nm. Concentration of red colorants were calibrated by the equation of dye(mg/ml) $A_{538nm}\times{1.284}$. Red colorants were changed to yellow at extreme alkali pH and repaired 55% color intensity by neutralization of pH and stabled below $55^\circ{C}$. Dyeability of red colorants against wool fabrics was mainly operated by red pigment having 538nm absorbance without big color differences. Below $55^\circ{C}$, color differences $(\Delta{E}^*_{ab})$ were not changed in spite of big difference of chroma$(c^*)$, having higher scores at higher temperature. The effect of mordants were not drastically changed parameters of color difference without copper ion. Citric acid was big changes of color difference$(\Delta{E}^*_{ab})$ in spite of similar chroma$(c^*)$ values. From these experimental results, red colorants from Phytolacca americana Linne is available for wool fabric dyeing.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2003년도 하계종합학술대회 논문집 Ⅳ
• /
• pp.2259-2262
• /
• 2003

This paper describes visual tracking procedure of the underwater mobile robot for nuclear reactor vessel inspection, which is required to find the foreign objects such as loose parts. The yellowish underwater robot body tend to present a big contrast to boron solute cold water of nuclear reactor vessel, tinged with indigo by Cerenkov effect. In this paper, we have found and tracked the positions of underwater mobile robot using the two color informations, yellow and indigo. The center coordinates extraction procedures is as follows. The first step is to segment the underwater robot body to cold water with indigo background. From the RGB color components of the entire monitoring image taken with the color CCD camera, we have selected the red color component. In the selected red image, we extracted the positions of the underwater mobile robot using the following process sequences: binarization labelling, and centroid extraction techniques. In the experiment carried out at the Youngkwang unit 5 nuclear reactor vessel, we have tracked the center positions of the underwater robot submerged near the cold leg and the hot leg way, which is fathomed to 10m deep in depth.

• International Journal of Internet, Broadcasting and Communication
• /
• 제16권2호
• /
• pp.22-30
• /
• 2024

Image compression-decompression methods have become increasingly crucial in modern times, facilitating the transfer of high-quality images while minimizing file size and internet traffic. Historically, early image compression relied on rudimentary codecs, aiming to compress and decompress data with minimal loss of image quality. Recently, a novel compression framework leveraging colorization techniques has emerged. These methods, originally developed for infusing grayscale images with color, have found application in image compression, leading to colorization-based coding. Within this framework, the encoder plays a crucial role in automatically extracting representative pixels-referred to as color seeds-and transmitting them to the decoder. The decoder, utilizing colorization methods, reconstructs color information for the remaining pixels based on the transmitted data. In this paper, we propose a novel approach to image compression, wherein we decompose the compression task into grayscale image compression and colorization tasks. Unlike conventional colorization-based coding, our method focuses on the colorization process rather than the extraction of color seeds. Moreover, we employ the Denoising Diffusion Null-Space Model (DDNM) for colorization, ensuring high-quality color restoration and contributing to superior compression rates. Experimental results demonstrate that our method achieves higher-quality decompressed images compared to standard JPEG and JPEG2000 compression schemes, particularly in high compression rate scenarios.

• 원예과학기술지
• /
• 제33권4호
• /
• pp.559-565
• /
• 2015

The rind phenotype of watermelon fruits is an important agronomic characteristic in the watermelon market. Inheritance and linkage analyses were performed for three rind-related traits that together determine the rind phenotype: foreground stripe pattern, rind background color, and depth of rind color. The inheritance of the foreground stripe pattern was analyzed using three different $F_2$ populations, showing that the striped pattern is dominant over the non-striped pattern. The inheritance analysis of the rind background color was performed using $F_2$ populations of the '10909' and '109905', and the depth of rind color was analyzed using $F_2$ populations of the '90509' and '109905'. Yellow color was found to be dominant over green color, and a deep color was dominant over the standard color. Linkage analysis of the three traits was conducted using three $F_2$ populations in which two traits were segregating. Each pair of traits was inherited independently, which demonstrated that the three traits are not linked. Therefore, we propose a three-locus model for the determination of rind phenotype, providing novel insight that rind phenotype is determined by the combination of three genetically independent loci.