• Korean Journal of Remote Sensing
• /
• v.35 no.2
• /
• pp.229-239
• /
• 2019

All vegetation colonies have layered structure. This layer is called 'forest vertical structure.' Nowadays it is considered as an important indicator to estimate forest's vital condition, diversity and environmental effect of forest. So forest vertical structure should be surveyed. However, vertical structure is a kind of inner structure, so forest surveys are generally conducted through field surveys, a traditional forest inventory method which costs plenty of time and budget. Therefore, in this study, we propose a useful method to classify the vertical structure of forests using remote sensing aerial photographs and machine learning capable of mass data mining in order to reduce time and budget for forest vertical structure investigation. We classified it as SVM (Support Vector Machine) using RGB airborne photos and LiDAR (Light Detection and Ranging) DSM (Digital Surface Model) DTM (Digital Terrain Model). Accuracy based on pixel count is 66.22% when compared to field survey results. It is concluded that classification accuracy of layer classification is relatively high for single-layer and multi-layer classification, but it was concluded that it is difficult in multi-layer classification. The results of this study are expected to further develop the field of machine learning research on vegetation structure by collecting various vegetation data and image data in the future.

• Archives of design research
• /
• v.17 no.2
• /
• pp.393-402
• /
• 2004

The mobile advertisement has varied on lifestyle of people, who live aninformation-oriented society with portable equipment such as web phone and PDA(Personal Digital Assistant). Also, the advertising has expended mobile techniques and its application field unpredictably. The intrinsic characteristic of misdistribution, reach, and convenience in mobile advertisement add up the capacity of a location, information and individualize. This market condition leads the basic audio focused formal mobile advertisement to the new mobile Internet environment with an additional able of data communication. Moreover, the type of SMS (Short Message Service), Graphic, Wep Push, and ridchmedia, which based on music, basic graphic, voice, and letters transfer by mobile terminal and the mobile character is present inevitably correlation with pixel art and animation in 2D(Two Dimensions) techniques. Thus, this research appoints the importance and its role of mobile advertisement that is core of the business marketing in new media era. To activate mobile market, the mobile companies classify the characteristic of consumers with developed commercial use of mobile character and research their behavior to meet optimal mobile character in business.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.6
• /
• pp.227-233
• /
• 2019

Recently, along with the recent development of deep learning technique, neural networks are achieving success in computer vision filed. Convolutional neural network have shown outstanding performance in not only for a simple image classification task, but also for tasks with high difficulty such as object segmentation and detection. However many such deep learning models are based on supervised-learning, which requires more annotation labels than image-level label. Especially image semantic segmentation model requires pixel-level annotations for training, which is very. To solve these problems, this paper proposes a weakly-supervised semantic segmentation method which requires only image level label to train network. Existing weakly-supervised learning methods have limitations in detecting only specific area of object. In this paper, on the other hand, we use multi-classifier deep learning architecture so that our model recognizes more different parts of objects. The proposed method is evaluated using VOC 2012 validation dataset.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.36 no.8C
• /
• pp.505-515
• /
• 2011

This paper describes a normalization method of Haar-like features used for object detection. Previous method which performs variance normalization on Haar-like features requires a lot of calculations, since it uses an additional integral image for calculating the standard deviation of intensities of pixels in a candidate window and increases possibility of false detection in the area where variance of brightness is small. The proposed normalization method can be performed much faster than the previous method by not using additional integral image and classifiers which are trained with the proposed normalization method show robust performance in various lighting conditions. Experimental result shows that the object detector which uses the proposed method is 26% faster than the one which uses the previous method. Detection rate is also improved by 5% without increasing false alarm rate and 45% for the samples whose brightness varies significantly.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.5
• /
• pp.141-147
• /
• 2021

In this paper, in order to improve the utilization of hyperspectral large-capacity data feature information by reducing complex computations by dimension reduction of neural network inputs in embedded systems, the band selection algorithm is applied in each subset. Among feature extraction and feature selection techniques, the feature selection aim to improve the optimal number of bands suitable for datasets, regardless of wavelength range, and the time and performance, more than others algorithms. Through this experiment, although the time required was reduced by 1/3 to 1/9 times compared to the others band selection technique, meaningful results were improved by more than 4% in terms of performance through the K-neighbor classifier. Although it is difficult to utilize real-time hyperspectral data analysis now, it has confirmed the possibility of improvement.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.13 no.6
• /
• pp.1353-1364
• /
• 2018

This study proposed a method to detect Cochlodinium polykrikoides red tide pixels in satellite images using a logistic regression model of machine learning technique under Imbalanced data. The spectral profiles extracted from red tide, clear water, and turbid water were used as training dataset. 70% of the entire data set was extracted and used for as model training, and the classification accuracy of the model was evaluated using the remaining 30%. At this time, the white noise was added to the spectral profile of the red tide, which has a relatively small number of data compared to the clear water and the turbid water, and over-sampling was performed to solve the unbalanced data problem. As a result of the accuracy evaluation, the proposed algorithm showed about 94% classification accuracy.

• Journal of Digital Convergence
• /
• v.20 no.4
• /
• pp.491-498
• /
• 2022

Based on Goethe's color theory, this study presented concepts and classification methods through the following research methods for the purpose of defining sensory, emotional, and experiential colors as spatial colors in digital space. First, the concept of spatial color is defined through theoretical consideration and three (3) types of spatial color are classified as the surface, outline, and physical colors. Secondly, the study includes the characteristics of digital space and color sensory type. Third, based on the identified color sensory type through the previous theoretical consideration, the four (4) categorized digital spatial color were derived and presented as techno Chromatic, S.E.N.S.E, pixel, and blur colors were determined and proposed. Based on such research contents, this study is meaningful in that it systematized the meaning of Goethe's color theory in the present age through digital spatial color.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.05a
• /
• pp.363-365
• /
• 2018

대장과 소장에서 모두 폐색 영역을 검출하기 위하여 본 논문에서는 기존에 연구된 장 폐색 영역 검출 방법과 ART2 알고리즘을 이용한 대장 폐색 영역과 소장 폐색 영역을 검출하는 방법을 제안한다. 제안된 방법은 기존에 연구된 방법을 이용하여 ROI 영역을 추출한 후, 추출된 ROI 영역을 ART2 알고리즘을 이용하여 영상을 군집화 한다. 군집화된 ROI 영역과 기존에 연구된 방법으로 X-ray 영상에서 검출한 장 폐색 영역의 형태학적 특징을 비교 및 분석하여 장 폐색의 형태학적 특징을 포함하는 클러스터를 분석한다. 따라서 장 폐색 영역에 해당되는 클러스터로 분류된 영역 내부를 클러스터의 중심에 해당되는 픽셀로 모두 대체한다. 그리고 $3^*3$ 필터를 이용한 침식과 팽창 연산을 적용하여 잡음을 제거한다. 잡음이 제거된 영상에서 각 객체들을 라벨링한 후에 크기를 비교하여 배경과 기타 지방 영역을 제거하고 남은 객체들을 장 폐색 영역으로 검출한다. 제안된 추출 방법을 장 폐색 X-ray 영상을 대상으로 실험한 결과, 기존에 연구된 방법으로 추출에 성공한 대장 장 폐색 영상과 추출에 실패한 소장 폐색 영상 모두에서 추출되는 것을 확인하였다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2013.11a
• /
• pp.223-226
• /
• 2013

정확한 움직임 추정 기술은 원본과 가장 유사한 영상의 복원에 효과적이고 압축률에도 중요한 영향을 미친다. 하지만 기존의 전역 탐색 (Full Search) 알고리듬과 Sum of Absolute Difference (SAD)라는 정합 오차 기준은 연산량이 높고 하드웨어 구현시 비효율적이다. 이를 보완하기 위한 1비트 변환 알고리듬은 움직임 벡터의 변화량을 0과 1의 연산으로 나타내는데, 이 알고리듬은 움직임이 많아 픽셀 값의 변화가 심한 블록의 변화량도 0과 1로만 표현한다. 그렇기 때문에 정확한 움직임이 반영되지 않고 그로 인해 낮은 Peak Siganl to Noise Ratio (PSNR)을 가져온다. 이 점을 개선하고자 본 논문에서는 참조블록들의 움직임 벡터를 파악하고 분류하여 선택적으로 움직임의 변화량이 큰 영역은 전역 탐색 알고리듬을, 움직임이 작거나 없는 영역은 1비트 연산을 수행하도록 하여 기존의 알고리듬과 비교하여 Peak Siganl to Noise Ratio (PSNR)측면에서 우수한 성능을 확인할 수 있었다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2007.10a
• /
• pp.985-988
• /
• 2007

This paper presents techniques to make image enhancement using unsharp masking. It is the technique to make image enhancement by automatically find the three parameters that makes hard to use the unsharp mask technique. To optimize the three parameters(Threshold, Amount, Radius), at first classify the pixels in the image to three groups, and then according to the groups, apply the unsharp mask to the image differently. We experimented and analyzed the rate of image enhancement by comparing images which is enhanced by human and which is enhanced by proposed technique.