• Journal of the Korea Institute of Military Science and Technology
• /
• v.25 no.6
• /
• pp.572-579
• /
• 2022

The analysis of foreign aircraft appearing suddenly in air defense identification zones requires a lot of cost and time. This study aims to develop a pre-trained model that can identify neighboring military aircraft based on aircraft photographs available on the web and present a model that can determine which aircraft corresponds to based on aerial photographs taken by allies. The advantages of this model are to reduce the cost and time required for model classification by proposing a pre-trained model and to improve the performance of the classifier by data augmentation of edge-detected images, cropping, flipping and so on.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2013.06a
• /
• pp.354-356
• /
• 2013

ITU와 ISO/IEC가 공동으로 UHD급 영상 부호화를 위해 표준화를 진행하고 있는 HEVC 코덱은 H.264/AVC 대비 2배 이상의 압축 효율을 갖는 것을 목표로 정하고 있다. HEVC(High Efficiency Video Coding)는 In-Loop Filter 기술로 H.264/AVC에서 사용하고 있는 Deblocking Filter와 새롭게 추가 된 SAO(Sample Adaptive Offset)를 사용하고 있다. 본 논문에서는 HEVC의 In-Loop Filter 기술 중 하나인 SAO의 기술의 EO에서 Category를 조금 더 정확하게 판단하여 분류하는 방법을 제안을 한다.

• Bulletin of the Korean Mathematical Society
• /
• v.59 no.1
• /
• pp.203-212
• /
• 2022

A simple graph is called semi-symmetric if it is regular and edge transitive but not vertex transitive. In this paper we prove that there is no connected cubic semi-symmetric graph of order 12p³ for any prime number p.

• Proceedings of the IEEK Conference
• /
• 2002.07b
• /
• pp.1007-1010
• /
• 2002

In this paper, we present a new blocking artifact reduction algorithm using interpolation and signal adaptive filter (SAF) based on the edge map. Generally, block-based coding, such as JPEG and MPEG, is the most popular image compression method. However, for high compression it produces noticeable blocking and ringing artifacts in the decoded image. In proposed method, all the block is classified into low and high frequency blocks in block classification procedure. And edge map is obtained by using Sobel operator on decoded image. And according to the block property we applied blocking artifacts reduction algorithm. Namely, four neighbor low frequency block is participated in interpolation based on edge map. And ringing artifacts is removed by applying a signal adaptive filter around the edge using edge map in high frequency block. The computer simulation results confirmed a better performance by the proposed method in both the subjective and objective image qualities.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.18 no.3
• /
• pp.89-100
• /
• 2023

Recently, on-device artificial intelligence (AI) solutions using mobile devices and embedded edge devices have emerged in various fields, such as computer vision, to address network traffic burdens, low-energy operations, and security problems. Although vision transformer deep learning models have outperformed conventional convolutional neural network (CNN) models in computer vision, they require more computations and parameters than CNN models. Thus, they are not directly applicable to embedded edge devices with limited hardware resources. Many researchers have proposed various model compression methods or lightweight architectures for vision transformers; however, there are only a few studies evaluating the effects of model compression techniques of vision transformers on performance. Regarding this problem, this paper presents a performance evaluation of vision transformers on embedded platforms. We investigated the behaviors of three vision transformers: DeiT, LeViT, and MobileViT. Each model performance was evaluated by accuracy and inference time on edge devices using the ImageNet dataset. We assessed the effects of the quantization method applied to the models on latency enhancement and accuracy degradation by profiling the proportion of response time occupied by major operations. In addition, we evaluated the performance of each model on GPU and EdgeTPU-based edge devices. In our experimental results, LeViT showed the best performance in CPU-based edge devices, and DeiT-small showed the highest performance improvement in GPU-based edge devices. In addition, only MobileViT models showed performance improvement on EdgeTPU. Summarizing the analysis results through profiling, the degree of performance improvement of each vision transformer model was highly dependent on the proportion of parts that could be optimized in the target edge device. In summary, to apply vision transformers to on-device AI solutions, either proper operation composition and optimizations specific to target edge devices must be considered.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.1738-1741
• /
• 1997

This paper addresses the issue fo target classification and localization with a SONAR for mobiler robot indoor navigation. In particular, multiple refetions of SONAR sound are used actively and interntionally. As for the SONAR sensor, the multiple reflection has been generally considered as one of the noisy phenomena, which is inevitable in the indoor environments. However, these multiple reflections can be a clue for classifying and localizing targets in the indoor environment if those can be controlled and used well. This paper develops a new SONAR sensor module with a reflection plane which can actively create the multiple refection. This paper also intends to suggest a new target classification emthod which uses the multiple refectiions. We approximate the world as being two dimensional and assume that the targets consisting of the indoor environment are pland, corner, and edge. Multiple reflection paths of an acoustic bean by a SONAR are analyzed, by simulations and the patterns of the TOPs (Time Of Flight) and angles of multiple reflections from each target are also analyzed. In addition, a new algorithm for target classification and localization is proposed.

• International Journal of Computer Science & Network Security
• /
• v.22 no.4
• /
• pp.101-110
• /
• 2022

Histopathological analysis of biopsy specimens is still used for diagnosis and classifying the brain tumors today. The available procedures are intrusive, time consuming, and inclined to human error. To overcome these disadvantages, need of implementing a fully automated deep learning-based model to classify brain tumor into multiple classes. The proposed CNN model with an accuracy of 92.98 % for categorizing tumors into five classes such as normal tumor, glioma tumor, meningioma tumor, pituitary tumor, and metastatic tumor. Using the grid search optimization approach, all of the critical hyper parameters of suggested CNN framework were instantly assigned. Alex Net, Inception v3, Res Net -50, VGG -16, and Google - Net are all examples of cutting-edge CNN models that are compared to the suggested CNN model. Using huge, publicly available clinical datasets, satisfactory classification results were produced. Physicians and radiologists can use the suggested CNN model to confirm their first screening for brain tumor Multi-classification.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.5
• /
• pp.48-55
• /
• 2009

Prior research works in indoor-outdoor classification have been conducted based on a simple combination of low-level features. However, since there are many challenging problems due to the extreme variability of the scene contents, most methods proposed recently tend to combine the low-level features with high-level information such as the presence of trees and sky. To extract these regions from videos, we need to conduct additional tasks, which may yield the increasing number of feature dimensions or computational burden. Therefore, an efficient indoor-outdoor scene classification method is proposed in this paper. First, the video is divided into the five same-sized blocks. Then we define and use the edge and color orientation histogram (ECOH) descriptors to represent each sub-block efficiently. Finally, all ECOH values are simply concatenated to generated the feature vector. To justify the efficiency and robustness of the proposed method, a diverse database of over 1200 videos is evaluated. Moreover, we improve the classification performance by using different weight values determined through the learning process.

• Journal of Intelligence and Information Systems
• /
• v.15 no.2
• /
• pp.1-13
• /
• 2009

Scene classification and concept-based procedures have been the great interest for image categorization applications for large database. Knowing the category to which scene belongs, we can filter out uninterested images when we try to search a specific scene category such as beach, mountain, forest and field from database. In this paper, we propose an adaptive segmentation method for real-world natural scene classification based on a semantic modeling. Semantic modeling stands for the classification of sub-regions into semantic concepts such as grass, water and sky. Our adaptive segmentation method utilizes the edge detection to split an image into sub-regions. Frequency of occurrences of these semantic concepts represents the information of the image and classifies it to the scene categories. K-Nearest Neighbor (k-NN) algorithm is also applied as a classifier. The empirical results demonstrate that the proposed adaptive segmentation method outperforms the Vogel and Schiele's method in terms of accuracy.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.8 no.6
• /
• pp.122-128
• /
• 1998

This paper presents compression of image sequences based on the classification of interframe difference image blocks. classification process consists of image activity classification and energy distribution classification. In the activity classification, interframe difference image blocks are classified into activity blocks and non-activity blocks using the edge detection. In the distribution classification, activity blocks are further classified into vertical blocks, horizontal blocks, and small activity blocks using the AC energy distribution features. The RBFN, trained with numerical classification results, successfully classifies difference image blocks according to image details. Image sequence compressing based on the classification of interframe difference image blocks using the RBFN shows better compression results and less training time than the classical sorting method and the MLP network.