• Journal of the Korean Society of Costume
• /
• v.65 no.5
• /
• pp.88-102
• /
• 2015

The purpose of this study is to investigate the examples of using illustration as a strategic factor of composing brand cultures and as a factor for brand identity design. Through analyzing the external characteristics and the internal characteristics of illustration, this study will give suggestions on ways to apply the examples to real design. Illustration in external characteristics is investigated as a case of applying it directly to fashion design and to fashion marketing. Most of the fashion items were printed or weaved and most of the bags, shoes, and accessories were printed on the cover, attached as a patch, and expressed three-dimensionally. Illustration in internal characteristics is investigated as fixing and expansion of brand image, improving artistic and emotional value of brand, vitalization of masstige items, and cultural support and expression of social responsibility. The three themes used to develop the illustrations of "Hello ZIBI", which was used in this study, were "Graphic", "Forest" and "Flower", and these were based on modified brand symbol. Casual brands grafted fashion item designs onto T-shirts, bags, hats, and scarves. Marketing items were designed as shopping bags that could reflect brand image, as well as other items, such as key holders, mug cups, and tumblers, with designs that targeted specific age groups.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.510-512
• /
• 2023

In order to enhance a model's capability for detecting facial expressions, this research suggests a pipeline that makes use of the GradCAM component. The patching module and the pseudo-labeling module make up the pipeline. The patching component takes the original face image and divides it into four equal parts. These parts are then each input into a 2Dconvolutional layer to produce a feature vector. Each picture segment is assigned a weight token using GradCAM in the pseudo-labeling module, and this token is then merged with the feature vector using principal component analysis. A convolutional neural network based on transfer learning technique is then utilized to extract the deep features. This technique applied on a public dataset MMI and achieved a validation accuracy of 96.06% which is showing the effectiveness of our method.

• Proceedings of the Korean Society of Crop Science Conference
• /
• 2022.10a
• /
• 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.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2017.06a
• /
• pp.59-62
• /
• 2017

본 논문에서는 평균 밝기 부분에 가중치 맵으로써 감마 변환에 기반한 선형 결합을 제안하고자 한다. 기존의 패치를 기반으로 한 가중치 맵은 평균 밝기 부분에서 영상 내 밝기 값이 한쪽으로 치우쳐 영상의 밝은 부분이 과포화 상태가 되어 세부 정보가 손실되는 단점이 있다. 이에 본 논문에서는 전역적 및 지역적 영상의 평균 밝기 값을 이용하여 감마 변환된 값을 선형 결합 시켜줌으로써 영역 내 세부 정보를 보존시키고 주관적 화질을 향상시켰다. 실험을 통해 결과를 분석하고 성능을 비교하여 기존 알고리듬에 비해 제안한 알고리듬이 우수함을 증명하였다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.11a
• /
• pp.95-98
• /
• 2022

아날로그 게이지는 여전히 많은 산업 시설에서 사용되고 있지만, 게이지 값을 사람이 수동으로 읽기 때문에 정확히 측정하기 위해 많은 시간이 소모가 되는 문제점이 있다. 이러한 이유로 최근에는 합성곱 신경망을 사용하여 아날로그 게이지 값을 자동으로 인식하는 연구가 진행되고 있다. 그러나 대부분의 선행연구들은 게이지가 촬영된 영상을 그대로 입력으로 사용하고 있으며, 이러한 방법은 사람이 게이지를 읽는 과정을 고려하였을 때 불필요한 부분이 많다. 본 논문에서는 게이지 전체 이미지를 학습에 사용하지 않고, 게이지의 특정 이미지 패치 기반으로 아날로그 게이지 값을 인식하는 방법을 제안한다. 제안하는 방법은 게이지의 중심, 눈금의 최소, 최대, 지침의 좌표를 기반으로 이미지 패치를 생성하고 채널 축으로 병합하여 학습을 진행하였으며, 최종적으로게이지의 각도를 계산한다. 이는 게이지의 평균 각도 오차를 통해 제안한 방법이 게이지 값을 인식하는데 우수한 성능이 보였으며, 게이지 이미지에 장애물이 있는 경우에도 게이지 값을 인식할 수 있음을 확인하였다.

• Journal of KIISE:Computer Systems and Theory
• /
• v.34 no.1
• /
• pp.38-48
• /
• 2007

In this paper, we propose an image-based method for modeling 3D objects with curved surfaces based on the NURBS (Non-Uniform Rational B-Splines) representation. Starting from a few calibrated images, the user specifies the corresponding curves by means of an interactive user interface. Then, the 3D curves are reconstructed using stereo reconstruction. In order to fit the curves easily using the interactive user interface, NURBS curves and surfaces are employed. The proposed surface modeling techniques include surface building methods such as bilinear surfaces, ruled surfaces, generalized cylinders, and surfaces of revolution. In addition to these methods, we also propose various advanced surface modeling techniques, including skinned surfaces, swept surfaces, and boundary patches. Based on these surface modeling techniques, it is possible to build various types of 3D shape models with textured curved surfaces without much effort. Also, it is possible to reconstruct more realistic surfaces by using proposed view-dependent texture acquisition algorithm. Constructed 3D shape model with curves and curved surfaces can be exported in VRML format, making it possible to be used in different 3D graphics softwares.

• Journal of Korea Multimedia Society
• /
• v.11 no.2
• /
• pp.153-162
• /
• 2008

We propose a novel image inpainting method composed of two parts: band matching and seamless cloning. In band matching, a band enclosing the boundary of a missing region is compared to those from the other parts of the image. The inner area of the minimum difference band is then copied to the missing region. Even though this band matching results in successful inpainting in many practical applications, brightness discontinuity (a seam) may appear between the filled missing region and its neighborhood. We apply seamless cloning to remove such discontinuity between the two regions. However, since this basic method using one patch may not deal with cases where there are abrupt changes of color or brightness along the boundary, we furthermore devise one more step: target sub-division. The target area is subdivided into small sub-areas, and the band matching and seamless cloning are applied to each of them. The multiple results from the sub-division are then ordered according to inpainting quality, which is measured based on the edge map or discontinuity map along the boundary band.

• Smart Media Journal
• /
• v.12 no.1
• /
• pp.9-16
• /
• 2023

Recently, a convolutional neural network (CNN) based system is being developed to overcome the limitations of human resources in the apple quality classification of farmhouse. However, since convolutional neural networks receive only images of the same size, preprocessing such as sampling may be required, and in the case of oversampling, information loss of the original image such as image quality degradation and blurring occurs. In this paper, in order to minimize the above problem, to generate a image patch based graph of an original image and propose a random walk-based positional encoding method to apply the graph transformer model. The above method continuously learns the position embedding information of patches which don't have a positional information based on the random walk algorithm, and finds the optimal graph structure by aggregating useful node information through the self-attention technique of graph transformer model. Therefore, it is robust and shows good performance even in a new graph structure of random node order and an arbitrary graph structure according to the location of an object in an image. As a result, when experimented with 5 apple quality datasets, the learning accuracy was higher than other GNN models by a minimum of 1.3% to a maximum of 4.7%, and the number of parameters was 3.59M, which was about 15% less than the 23.52M of the ResNet18 model. Therefore, it shows fast reasoning speed according to the reduction of the amount of computation and proves the effect.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.22 no.10
• /
• pp.2112-2122
• /
• 1997

This paper presents a new video codec which is based on 2-D triangular mesh-based motion compensation and two step grid point motion estimation. With this approach the efficiency of compression and the quality of reconstructed images are improved. The compensation of motion for each triangular patch is performed by image warping using motion vectors at the grid points. The prediction error coding and the rate control meet MPEG-4 VM 3.0 specification. The experimental results show that the codec system proposed is simple in complexity and moreover, the quality of decoded images is improved.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.11 no.6
• /
• pp.283-288
• /
• 2011

This paper have proposed an object retrieval using the corners area variability based on correlogram. The proposed algorithm is processed as follows. First, the corner points of the object in an image are extracted and then the feature vectors are obtained. It are rearranged according to the number dimension and consist of sequence vectors. And the similarity based on the maximum of sequence vectors is measured. The proposed technique is invariant to the rotation or the transfer of the objects and more efficient in case that the objects present simple structure. In simulation that use Wang's database, the method presents that the recall property is improved by 0.03% and more than the standard corner patch histogram.