• Journal of KIISE:Databases
• /
• v.36 no.5
• /
• pp.366-373
• /
• 2009

In this paper, we propose a similarity search algorithm for image databases. One of the central problems regarding content-based image retrieval (CBIR) is the semantic gap between the low-level features computed automatically from images and the human interpretation of image content. Many search algorithms used in CBIR have used the Minkowski metric (or $L_p$-norm) to measure similarity between image pairs. However those functions cannot adequately capture the aspects of the characteristics of the human visual system as well as the nonlinear relationships in contextual information. Our new search algorithm tackles this problem by employing new similarity measures and ranking strategies that reflect the nonlinearity of human perception and contextual information. Our search algorithm yields superior experimental results on a real handwritten digit image database and demonstrates its effectiveness.

• Journal of KIISE
• /
• v.44 no.12
• /
• pp.1313-1318
• /
• 2017

Aggregating local features in a single vector is a fundamental problem in an image search. In this process, the image search process can be speeded up if binary features which are extracted almost two order of magnitude faster than gradient-based features are utilized. However, in order to utilize the binary features in an image search, it is necessary to study the techniques for clustering binary features to generate binary visual words. This investigation is necessary because traditional clustering techniques for gradient-based features are not compatible with binary features. To this end, this paper studies the techniques for clustering binary features for the purpose of generating binary visual words. Through experiments, we analyze the trade-off between the accuracy and computational efficiency of an image search using binary features, and we then compare the proposed techniques. This research is expected to be applied to mobile applications, real-time applications, and web scale applications that require a fast image search.

• Journal of Information Technology Services
• /
• v.10 no.4
• /
• pp.243-258
• /
• 2011

With increasing interest in ubiquitous computing technology, an infrastructure for the short-distance wireless communication has been extended socially, bringing spotlight to the security system using the image or location. In case of existing security system, there have been issues such as the occurrences of blind spots, difficulty in recognizing multiple objects and storing of the unspecified objects. In order to solve this issue, zone-based location-estimation search system for the image have been suggested as an alternative based on the real-time location determination technology combined with image. This paper intends to suggest the search service for the image zone-based location-estimation. For this, it proposed the location determination algorism using IEEE 802.15.4/ZigBee's RSSI and for real-time image service, the RTP/RTCP protocol was applied. In order to combine the location and image, at the event of the entry of the specified target, the record of the time for image and the time of occurrence of the event on a global time standard, it has devised a time stamp, applying XML based meta data formation method based on the media's feature data based in connection with the location based data for the events of the object. Using the proposed meta data, the service mode which can search for the image from the point in time when the entry of the specified target was proposed.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.7
• /
• pp.816-822
• /
• 2016

Content-Based image retrieval is a method to search by image features such as local color, texture, and other image content information, which is different from conventional tag or labeled text-based searching. In real life data, the number of images having tags or labels is relatively small, so it is hard to search the relevant images with text-based approach. Existing image search method only based on image feature similarity has limited performance and does not ensure that the results are what the user expected. In this study, we propose and validate a machine learning based approach to improve the performance of the image search engine. We note that when users search relevant images with a query image, they would expect the retrieved images belong to the same category as that of the query. Image classification method is combined with the traditional image feature similarity method. The proposed method is extensively validated on a public PASCAL VOC dataset consisting of 11,530 images from 20 categories.

• Journal of KIISE:Software and Applications
• /
• v.32 no.9
• /
• pp.853-867
• /
• 2005

A number of content-based image search methods have been proposed to this point. Each of these systems uses different image data and generates different data depending on the extraction method of different characteristics that the search capabilities of each system cannot be compared and assessed. In particular, there is a problem of applying the identical image data onto the contents based image search system on the web that cannot be compared and assessed. To resolve such a problem, the XML-based FEMAL is hereby presented for extracting data of characteristics generated from specific search system in a way that can be recognized from other starch system. In the experiment using FEMAL, the extract data for characteristics is mutually communicated and integrated and the comparison assessment of search capability is seemed to be available.

• Journal of the Semiconductor & Display Technology
• /
• v.6 no.2 s.19
• /
• pp.53-57
• /
• 2007

In this paper, a new content-based color image retrieval method is proposed, in which both the color content and the spatial relationship of image have been taken into account. In order to represent the spatial distribution information of image, a disorder matrix, which has the invariance to the rotation and translation of the image content, has been designed. This is based on multi-resolution color-spatial information. We present our algorithm in the following section, and then verified the search results with comparison to other methods, such as color histogram, wavelet histogram, correlogram and wavelet correlogram. Experimental results with various types of images show that the proposed method not only achieves a high image retrieval performance but also improve the retrieval precision.

• Journal of Sensor Science and Technology
• /
• v.30 no.5
• /
• pp.290-294
• /
• 2021

Experts have designed popular and successful model architectures, which, however, were not the optimal option for different scenarios. Despite the remarkable performances achieved by deep neural networks, manually designed networks for classification tasks are the backbone of object detection. One major challenge is the ImageNet pre-training of the search space representation; moreover, the searched network incurs huge computational cost. Therefore, to overcome the obstacle of the pre-training process, we introduce a network adaptation technique using a pre-trained backbone model tested on ImageNet. The adaptation method can efficiently adapt the manually designed network on ImageNet to the new object-detection task. Neural architecture search (NAS) is adopted to adapt the architecture of the network. The adaptation is conducted on the MobileNetV2 network. The proposed NAS is tested using SSDLite detector. The results demonstrate increased performance compared to existing network architecture in terms of search cost, total number of adder arithmetics (Madds), and mean Average Precision(mAP). The total computational cost of the proposed NAS is much less than that of the State Of The Art (SOTA) NAS method.

• Proceedings of the Korean Information Science Society Conference
• /
• 2010.06c
• /
• pp.62-65
• /
• 2010

Image annotations allow users to access a large image database with textual queries. But since the surrounding text of Web images is generally noisy. an efficient image annotation and retrieval system is highly desired. which requires effective image search techniques. Data mining techniques can be adopted to de-noise and figure out salient terms or phrases from the search results. Clustering algorithms make it possible to represent visual features of images with finite symbols. Annotationbased image search engines can obtains thousands of images for a given query; but their results also consist of visually noise. In this paper. we present a new algorithm Double-Circles that allows a user to remove noise results and characterize more precise representative annotations. We demonstrate our approach on images collected from Flickr image search. Experiments conducted on real Web images show the effectiveness and efficiency of the proposed model.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.3
• /
• pp.368-373
• /
• 2007

In recent year s there have been increasing interests in real-time object tracking with image information. This dissertation presents a real-time object tracking method through the object recognition based on neural networks that have robust characteristics under various illuminations. This dissertation proposes a global search and a local search method to track the object in real-time. The global search recognizes a target object among the candidate objects through the entire image search, and the local search recognizes and track only the target object through the block search. This dissertation uses the object color and feature information to achieve fast object recognition. The experiment result shows the usefulness of the proposed method is verified.

• Proceedings of the IEEK Conference
• /
• 2003.07e
• /
• pp.1851-1854
• /
• 2003

This paper suggests an algorithm which can retrieval images using correlations between the region classification of spatial image and the wavelet transform even though the images are rotated in a ${\pm}$90 degree arc. Owing to this proposed method, it was confirmed from experiments that the search about the whole image is not processed and only a few amount of informations are saved by using the mathematical statistics from the block map and transformed band which is resulted from region classification, and by performing the image search based on these, the improvement of search speed and the efficient search can be done.