• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.9B
• /
• pp.1588-1595
• /
• 2000

To restore image degraded by out-of-focus blur and additivie noise a regularized iterative restoration is used. In concentional method, regularization is usually applied to all over the image without considering the local characteristics of image. As a result, ringing artifacts appear in edge regions and the noise amplification is introduced in flat regions. To solve this problem we propose an adaptive regularization iterative restoration using directional regularization operator considering edges in four directions and the regularization operator with on direction for flat regions. We verified that the proposed method show better results in the suppression of the noise amplification in flat regions, and introduced less ringing artifacts in edge regions. As a result it showed visually better image and improved better ISNR further than the conventional methods.

• Journal of Convergence for Information Technology
• /
• v.7 no.2
• /
• pp.77-82
• /
• 2017

In this paper, a new security algorithm to recognize the type of the vehicle with the vehicle image as a input image is suggested. The vehicle recognition security algorithm is composed of five core parts, such as the input image, background removal, edge areas extraction, pre-processing(binarization), and the vehicle recognition. Therefore, the final recognition rate of the security algorithm for vehicle type recognition can be affected by the function and efficiency of each step. After inputting image into a gray scale image and removing backgrounds, the binarization is performed by extracting only the edge region. After the pre-treatment process for making outlines clear, the type of vehicles is categorized into large vehicles, passenger cars and motorcycles through the ratio of height and width of the vehicle.

• Journal of Magnetics
• /
• v.18 no.3
• /
• pp.311-316
• /
• 2013

Digital image processing has increasingly been implemented in nanostructural analysis and would be an ideal tool to characterize the morphology and position of self-assembled magnetic nanoparticles for high density recording. In this work, magnetic nanoparticles were synthesized by the modified polyol process using $Fe(acac)_3$ and $Pt(acac)_2$ as starting materials. Transmission electron microscope (TEM) images of as-synthesized products were inspected using an image processing procedure. Grayscale images ($800{\times}800$ pixels, 72 dot per inch) were converted to binary images by using Otsu's thresholding. Each particle was then detected by using the closing algorithm with disk structuring elements of 2 pixels, the Canny edge detection, and edge linking algorithm. Their centroid, diameter and area were subsequently evaluated. The degree of polydispersity of magnetic nanoparticles can then be compared using the size distribution from this image processing procedure.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.12
• /
• pp.3364-3382
• /
• 2023

Remote sensing image segmentation plays an important role in realizing intelligent city construction. The current mainstream segmentation networks effectively improve the segmentation effect of remote sensing images by deeply mining the rich texture and semantic features of images. But there are still some problems such as rough results of small target region segmentation and poor edge contour segmentation. To overcome these three challenges, we propose an improved semantic segmentation model, referred to as MRU-Net, which adopts the U-Net architecture as its backbone. Firstly, the convolutional layer is replaced by BasicBlock structure in U-Net network to extract features, then the activation function is replaced to reduce the computational load of model in the network. Secondly, a hybrid multi-scale recognition module is added in the encoder to improve the accuracy of image segmentation of small targets and edge parts. Finally, test on Massachusetts Buildings Dataset and WHU Dataset the experimental results show that compared with the original network the ACC, mIoU and F1 value are improved, and the imposed network shows good robustness and portability in different datasets.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.1
• /
• pp.137-143
• /
• 2017

X-ray image is a widely used to medical examination, airport security and cargo inspection. However, X-ray images contain many visual noise, which interrupt image analysis. Consequently, it is primary importance to reduce noises of X-ray image. In this paper, we present a improved denoise technique for x-ray image using pixel value and range weights. First, we denoise a x-ray image using bilateral filter. Next, we detect a edge region of the original x-ray image. If a denoised pixel belongs to the edge region, we calculate weighting values of original x-ray image and denoised x-ray image in $3{\times}3$ neighboring pixels and compute the cost value to determine the boundary pixel value. Finally, the pixel value having minimum cost is determined as the pixel value of the denoised x-ray image. Simulation results show that the proposed algorithm achieves good performance in terns of PSNR comparison and subjective visual quality.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.43 no.1 s.307
• /
• pp.27-34
• /
• 2006

In this paper, we propose a combined region-based image retrieval system using EHD(Edge Histogram Descriptor) and CLD(Color Layout Descriptor) of MPEG-7 descriptors. The combined descriptor can efficiently describe edge and color features in terms of sub-image regions. That is, the basic unit for the selection of the region-of-interest (ROI) in the image is the sub-image block of the EHD, which corresponds to 16 (i.e., $4{\times}4)$ non-overlapping image blocks in the image space. This implies that, to have a one-to-one region correspondence between ELE and CLD, we need to take an $8{\times}8$ inverse DCT (IDCT) for the CLD. Experimental results show that the proposed retrieval scheme can be used for image retrieval with the ROI based image retrieval for MPEG-7 indexed images.

• 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.

• Journal of Sensor Science and Technology
• /
• v.29 no.2
• /
• pp.133-140
• /
• 2020

This study proposes a method of blending visible and near infrared images to enhance edge details and local contrast. The proposed method consists of radiance map generation and color compensation. The radiance map is produced by a Laplacian pyramid and a soft mixing method based on principal component analysis. The color compensation method uses the ratio between the composed radiance map and the luminance channel of a visible image to preserve the visible image chrominance. The proposed method has better edge details compared to a conventional visible and NIR image blending method.

• Journal of the Korean Society for Precision Engineering
• /
• v.12 no.6
• /
• pp.72-79
• /
• 1995

The purpose of this paper is to develop an automatic measuring system based on the digital image processing which can be applied to the in-process measurment of the characteristics of the thin thickness. The derivative operators is used for edge detection in gray level image. This concept can be easily illustrated with the aid of object shows an image of a simple light object on a dark background, the gray level profile along a horizontal scan line of the image, and the first and second derivatives of the profile. The first derivative of an edge modeled in this manner is 0 in all regions of constant gray level, and assumes a constant value during a gray level transition. The experimental results indicate that the developed automatic inspection system can be applied in real situation.

• Proceedings of the KSRS Conference
• /
• 2008.10a
• /
• pp.126-129
• /
• 2008

In this paper, we proposed a method extracting an object from background of the satellite image. The image segmentation techniques have been widely studied for the technology to segment image and to synthesis segment object with other images. Proposed algorithm is to perform the edge detection of a selected object using genetic algorithm. We segment region of object based on detection edge using watershed algorithm. We separated background and object in indefinite region using gradual region merge from segment object. And, we make GUI for the application of the proposed algorithm to various tests. To demonstrate the effectiveness of the proposed method, several analysis on the satellite images are performed.