• Journal of Intelligence and Information Systems
• /
• v.15 no.2
• /
• pp.49-59
• /
• 2009

Contents-based image retrieval methods are in general more objective and effective than text-based image retrieval algorithms since they use color and texture in search and avoid annotating all images for search. SIM(Self-organizing Image browsing Map) is one of contents-based image retrieval algorithms that uses only browsable mapping results obtained by SOM(Self Organizing Map). However, SOM may have an error in selecting the right BMU in learning phase if there are similar nodes with distorted color information due to the intensity of light or objects' movements in the image. Such images may be mapped into other grouping nodes thus the search rate could be decreased by this effect. In this paper, we propose an improved SIM that uses HSV color model in extracting image features with color quantization. In order to avoid unexpected learning error mentioned above, our SOM consists of two layers. In learning phase, SOM layer 1 has the color feature vectors as input. After learning SOM Layer 1, the connection weights of this layer become the input of SOM Layer 2 and re-learning occurs. With this multi-layered SOM learning, we can avoid mapping errors among similar nodes of different color information. In search, we put the query image vector into SOM layer 2 and select nodes of SOM layer 1 that connects with chosen BMU of SOM layer 2. In experiment, we verified that the proposed SIM was better than the original SIM and avoid mapping error effectively.

• Journal of KIISE:Databases
• /
• v.35 no.4
• /
• pp.345-355
• /
• 2008

SIFT is popularly used in computer vision application such as object recognition, motion tracking, and 3D reconstruction among various shape descriptors. However, it is not easy to apply SIFT into the image similarity search as it is since it uses many high dimensional keypoint vectors. In this paper, we present a SIFT based image similarity search method using an edge image pyramid and an interesting region detection. The proposed method extracts keypoints, which is invariant to contrast, scale, and rotation of image, by using the edge image pyramid and removes many unnecessary keypoints from the image by using the hough transform. The proposed hough transform can detect objects of ellipse type so that it can be used to find interesting regions. Experimental results demonstrate that the retrieval performance of the proposed method is about 20% better than that of traditional SIFT in average recall.

• Proceedings of the IEEK Conference
• /
• 2000.09a
• /
• pp.363-366
• /
• 2000

이미지의 한가지 특징(feature)만을 고려한 내용 기반 이미지 검색(content-based image retrieval)은 두가지 이상의 특징 정보를 사용했을 경우와 비교하여 정확도(precision)가 떨어져 성능을 저하시킬 수 있다 따라서 대부분의 검색 시스템에서는 색상(color)이나 형태(shape), 질감(texture) 등과 같은 이미지의 다양한 특징들을 결합하여 검색에 이용하고 있다. 본 논문에서는 이미지의 색상 및 형태 정보를 이용하여 사용자의 질의와 유사한 이미지를 제공하고, 고 차원화된 이미지의 특징들을 클러스터링(clustering) 방법을 이용하여 빠르게 검색할 수 있도록 하였으며, 또한 검색시 그룹 경계 보정 방법을 이용하여 전체 검색을 하지 않고도 전체검색 결과와 동일한 결과를 얻을 수 있는 시스템을 설계 및 구현하였다. 실험에 사용된 데이터는 2022개의 자연 영상이였으며, HSI 색상 정보와 이미지의 에지(edge) 정보를 특징 벡터로 삼았다. 실험 결과, 색상 정보 하나만을 사용한 경우보다 정확도와 재현율면에서 사용자가 원하는 이미지와 보다 유사한 결과를 검출할 수 있었을 뿐만 아니라 클러스터링을 사용함으로써 보다 빠르고, 전체검색 결과와 동일한 검색이 가능하다는 것을 입증하였다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.10a
• /
• pp.429-430
• /
• 2018

The introduction of mobile devices increased the need to apply limitations such as mobile devices' memory, speed, energy, and bandwidth on the designs of searching images. There is a demand to reduce such limitations on searching images on the mobile phone. Hence, this paper proposes a design that adds tags on pictures to manage the images in mobile environment, allowing efficient searches and deletion of duplicate files based on the similarities of the images. The proposed method does not compromise its efficiency by increasing costs; it also reduces the volume of data needed for mobile devices.

• The Korean Journal of Applied Statistics
• /
• v.34 no.2
• /
• pp.191-204
• /
• 2021

Visual Question Answering (VQA) and image captioning are tasks that require understanding of the features of images and linguistic features of text. Therefore, co-attention may be the key to both tasks, which can connect image and text. In this paper, we propose a model to achieve high performance for VQA by image caption generated using a pretrained standard transformer model based on MSCOCO dataset. Captions unrelated to the question can rather interfere with answering, so some captions similar to the question were selected to use based on a similarity to the question. In addition, stopwords in the caption could not affect or interfere with answering, so the experiment was conducted after removing stopwords. Experiments were conducted on VQA-v2 data to compare the proposed model with the deep modular co-attention network (MCAN) model, which showed good performance by using co-attention between images and text. As a result, the proposed model outperformed the MCAN model.

• The Journal of the Korea Contents Association
• /
• v.21 no.8
• /
• pp.1-9
• /
• 2021

Deep learning in computer vision has made accelerated improvement over a short period but large-scale learning data and computing power are still essential that required time-consuming trial and error tasks are involved to derive an optimal network model. In this study, we propose a similar image classification performance improvement method based on CR (Confusion Rate) that considers only the characteristics of the data itself regardless of network optimization or data reinforcement. The proposed method is a technique that improves the performance of the deep learning model by calculating the CRs for images in a dataset with similar characteristics and reflecting it in the weight of the Loss Function. Also, the CR-based recognition method is advantageous for image identification with high similarity because it enables image recognition in consideration of similarity between classes. As a result of applying the proposed method to the Resnet18 model, it showed a performance improvement of 0.22% in HanDB and 3.38% in Animal-10N. The proposed method is expected to be the basis for artificial intelligence research using noisy labeled data accompanying large-scale learning data.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.1220-1223
• /
• 2023

폐기물을 처리하는데 있어 배출과 수거에 대한 프로세스 자동화를 위해 폐기물 객체 유사도 판별이 요구된다. 이를 위해 본 연구에서는 폐기물 데이터셋에서 SIFT(Scale-Invariant Feature Transform)와 HSV(Hue, Saturation, Value)기반으로 두 이미지의 공통된 특징을 추출해 융합하고, 기계학습을 통해 이미지 객체 간의 유사도를 측정하는 모델을 제안한다. 실험을 위해 수집된 폐기물 데이터셋 81,072 장을 활용하여 이미지를 학습시키고, 전통적인 임계치 기반 유사도 측정과 본 논문에서 제시하는 유사도 측정을 비교하여 성능을 확인하였다. 임계치 기반 측정에서 SIFT 와 HSV 는 각각 0.82, 0.89(Acc)가 측정되었고, 본 논문에서 제시한 특징 추출 방법을 사용한 기계학습의 성능은 DT(Decision Tree)와 SVM(Support Vector Machine) 모두 0.93 (Acc)로 4%의 정확도가 향상되었다.

• Smart Media Journal
• /
• v.9 no.4
• /
• pp.162-169
• /
• 2020

Design is a key factor that determines the competitiveness of products in the textile and fashion industry. It is very important to measure the similarity of the proposed design in order to prevent unauthorized copying and to confirm the originality. In this study, a deep learning technique was used to quantify features from images of textile designs, and similarity was measured using Spearman correlation coefficients. To verify that similar samples were actually detected, 300 images were randomly rotated and color changed. The results of Top-3 and Top-5 in the order of similarity value were measured to see if samples that rotated or changed color were detected. As a result, the VGG-16 model recorded significantly higher performance than did AlexNet. The performance of the VGG-16 model was the highest at 64% and 73.67% in the Top-3 and Top-5, where similarity results were high in the case of the rotated image. appear. In the case of color change, the highest in Top-3 and Top-5 at 86.33% and 90%, respectively.

• Proceedings of the IEEK Conference
• /
• 2000.09a
• /
• pp.393-396
• /
• 2000

본 논문에서는 영상의 화소값으로부터 추출된 유사 특징점(quasi-feature point)을 이용한 이미지 모자이킹 알고리즘을 제안한다. 유사 특징점의 선택은 전역 정합(global matching)의 결과로부터 중첩된 영역을 4개의 부영역(sub-area)으로 분할하고, 각각의 분할된 부 영역에서 국부 분산(local variance)의 크기가 큰 블록을 선정, 이 블록의 중심 화소를 유사 특징점으로 선택한다. 유사 특징점에 대한 정합은 카메라 이동에 따른 왜곡(distortion)과 조명의 변화를 고려한 블록 정합 알고리즘(block matching algorithm)을 이용한다.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.40 no.3
• /
• pp.171-180
• /
• 2003

A content-based image retrieval(CBIR) system builds the image database using low-level features such as color, shape and texture and provides similar images that user wants to retrieve when the retrieval request occurs. What the user is interest in is a response time in consideration of the building time to build the index database and the response time to obtain the retrieval results from the query image. In a content-based image retrieval system, the similarity computing time comparing a query with images in database takes the most time in whole response time. In this paper, we propose the two-phase search method with the clustering technique of feature vector in order to minimize the similarity computing time. Experimental results show that this two-phase search method is 2-times faster than the conventional full-search method using original features of ail images in image database, while maintaining the same retrieval relevance as the conventional full-search method. And the proposed method is more effective as the number of images increases.