Earticle

This paper designed and developed the image processing system of integrating feature extraction and matching by using convolutional neural network (CNN), rather than relying on the simple method of processing feature extraction and matching separately in the image processing of conventional image recognition system. To implement it, the proposed system enables CNN to operate and analyze the performance of conventional image processing system. This system extracts the features of an image using CNN and then learns them by the neural network. The proposed system showed 84% accuracy of recognition. The proposed system is a model of recognizing learned images by deep learning. Therefore, it can run in batch and work easily under any platform (including embedded platform) that can read all kinds of files anytime. Also, it does not require the implementing of feature extraction algorithm and matching algorithm therefore it can save time and it is efficient. As a result, it can be widely used as an image recognition program.

TRISO (Tri-Isotropic)-coated fuel particle is widely applied due to its higher stability at high temperature and its efficient retention capability for fission products in the HTGR (high temperature gas-cooled reactor), one of the highly efficient Generation IV reactors. The typical ball-type TRISO-coated fuel particle with a diameter of about 1 mm is composed of a nuclear fuel particle as a kernel and of outer coating layers. The coating layers consist of a buffer PyC, inner PyC, SiC, and outer PyC layer. In this study, a digital image processing algorithm is proposed to automatically measure the thickness of the coating layers. An FBP (filtered backprojection) algorithm was applied to reconstruct the CT image using virtual X-ray radiographic images for a simulated TRISO-coated fuel particle. The automatic measurement algorithm was developed to measure the coating thickness for the reconstructed image with noises. The boundary lines were automatically detected, then the coating thickness was circularly by the algorithm. The simulation result showed that the measurement error rate was less than 1.4%.

Vehicle detector system based on image processing technology is a significant domain of ITS (Intelligent Transportation System) applications due to its advantages such as low installation cost and it does not obstruct traffic during the installation of vehicle detection systems on the road[1]. In this paper, we propose architecture for vehicle detection by using image processing. The architecture consists of two main parts such as an image processing part, using high speed FPGA, decision and calculation part using CPU. The CPU part takes care of total system control and synthetic decision of vehicle detection. The FPGA part assumes charge of input and output image using video encoder and decoder, image classification and image memory control.

Accurate forest fires detection algorithms remain a challenging issue, because, some of the objects have the same features with fire, which may result in high false alarms rate. This paper presents a new video-based, image processing forest fires detection method, which consists of four stages. First, a background-subtraction algorithm is applied to detect moving regions. Secondly, candidate fire regions are determined using CIE $L{\ast}a{\ast}b{\ast}$ color space. Thirdly, special wavelet analysis is used to differentiate between actual fire and fire-like objects, because candidate regions may contain moving fire-like objects. Finally, support vector machine is used to classify the region of interest to either real fire or non-fire. The final experimental results verify that the proposed method effectively identifies the forest fires.

To ensure safety of bridges, it is critical to inspect and assess physical and functional conditions regularly. Presently, most highway bridges in the U.S. are inspected visually. However, this method of inspection is often influenced by the bridge inspector's knowledge and experience. So, reliability and accuracy of inspection results may be problematic. To solve such problems, an extensive number of robotics systems and image processing techniques for bridge inspection methods have been proposed. These robotics systems and image processing techniques are used to measure various bridge conditions, such as apparent damage, displacement and dynamic characteristics. This paper provides a comprehensive review of robotics systems and image processing technologies used in bridge inspection.

Corner detection and feature extraction are essential aspects of computer vision problems such as object recognition and tracking. Feature detectors such as Scale Invariant Feature Transform (SIFT) yields high quality features but computationally intensive for use in real-time applications. The Features from Accelerated Segment Test (FAST) detector provides faster feature computation by extracting only corner information in recognising an object. In this paper we have analyzed the efficient object detection algorithms with respect to efficiency, quality and robustness by comparing characteristics of image detectors for corner detector and feature extractors. The simulated result shows that compared to conventional SIFT algorithm, the object recognition system based on the FAST corner detector yields increased speed and low performance degradation. The average time to find keypoints in SIFT method is about 0.116 seconds for extracting 2169 keypoints. Similarly the average time to find corner points was 0.651 seconds for detecting 1714 keypoints in FAST methods at threshold 30. Thus the FAST method detects corner points faster with better quality images for object recognition.

It is important to understand the exact habitat distribution of endangered species because of their decreasing numbers. In this study, we build a system with a deep learning module that collects the image data of endangered animals, processes the data, and saves the data automatically. The system provides a more efficient way than human effort for classifying images and addresses two problems faced in previous studies. First, specious answers were suggested in those studies because the probability distributions of answer candidates were calculated even if the actual answer did not exist within the group. Second, when there were more than two entities in an image, only a single entity was focused on. We applied an object detection algorithm (YOLO) to resolve these problems. Our system has an average precision of 86.79%, a mean recall rate of 93.23%, and a processing speed of 13 frames per second.

Text-based image retrieval is not only cumbersome as it requires the manual input of keywords by the user, but is also limited in the semantic approach of keywords. However, content-based image retrieval enables visual processing by a computer to solve the problems of text retrieval more fundamentally. Vision applications such as extraction and mapping of image characteristics, require the processing of a large amount of data in a mobile environment, rendering efficient power consumption difficult. Hence, an effective image retrieval method on mobile platforms is proposed herein. To provide the visual meaning of keywords to be inserted into images, the efficiency of image retrieval is improved by extracting keywords of exchangeable image file format metadata from images retrieved through a content-based similar image retrieval method and then adding automatic keywords to images captured on mobile devices. Additionally, users can manually add or modify keywords to the image metadata.

Three-dimensional (3D) medical image segmentation is vital in clinical applications, but faces challenges like computational complexity and spatial context loss in 2D processing. To address these issues, we propose 3DAS2D, which processes 3D images as 2D slices while retaining volumetric data for better accuracy. Our approach integrates modules like Neighbour Attention, Memory Attention, Memory Bank, and Mask Encoder to capture inter-slice dependencies and historical context within a 2D prediction framework. Evaluated on four datasets covering prostate, cardiac, and lung tumour tasks, 3DAS2D performed comparably to 3D methods. This work offers an efficient, adaptable solution bridging 2D efficiency and 3D accuracy, with significant clinical potential.

This paper proposes a technique for principal component analysis (PCA) to raise the recognition rate of a front face in a low dimension by hierarchical image and parallel processing structure. The conventional PCA shows a recognition rate of less than 50% in a low dimension (dimensions 1 to 6) when used for facial recognition. In this paper, a face is formed as images of 3 fixed-size levels: the 1st being a region around the nose, the 2nd level a region including the eyes, nose, and mouth, and the 3rd level image is the whole face. PCA of the 3-level images is treated by parallel processing structure, and finally their similarities are combined for high recognition rate in a low dimension. The proposed method under went experimental feasibility study with ORL face database for evaluation of the face recognition function. The experimental demonstration has been done by PCA and the proposed method according to each level. The proposed method showed high recognition of over 50% from dimensions 1 to 6.

Color image processing is very useful because color is a strong marker that also simplifies the recognition and extraction of objects from a scene and humans can distinguish thousands of shades of color and intensity compared to just around two dozen shades of grey. The processing of color images can be classified into two main areas as full-color and pseudo-colour. The images are first acquired with a full color sensor such as color TV camera, color scanner and second category has the question of assigning a color to a different monochrome intensity and intensity ranges. The aim of a color model is to promote color specification in a widely agreed way to some standard. A color model is a coordinate system specification and a subspace within that system where a single point represents each color. Some particular conversions can convert different color models to one another. This paper specify overview of different types of images, color models, their conversion and applications.

This paper proposes a gradation image processing method for the development of a Road Sign Recognition Platform (RReP), which aims to facilitate the rapid and accurate management and surveying of approximately 160,000 road signs installed along the highways, national roadways, and local roads in the cities, districts (gun), and provinces (do) of Korea. RReP is based on GPS(Global Positioning System), IMU(Inertial Measurement Unit), INS(Inertial Navigation System), DMI(Distance Measurement Instrument), and lasers, and uses an imagery information collection/classification module to allow the automatic recognition of signs, the collection of shapes, pole locations, and sign-type data, and the creation of road sign registers, by extracting basic data related to the shape and sign content, and automated database design. Image division and merging, which were applied in this study, produce superior results compared with local binarization method in terms of speed. At the results, larger texts area were found in images, the accuracy of text recognition was improved when images had been gradated. Multi-threshold values of natural scene images are used to improve the extraction rate of texts and figures based on pattern recognition.

본 논문은 이미지 처리 기술을 이용한 광 카메라 통신(OCC: Optical Camera Communications) 기술을 제안한다. OCC 시스템의구조 및 동작을 제안한다. 상용 30fps 카메라의 샘플링 동작에 의해 제한되는 데이터율을 증가시키기 위해 칼라 이미지 처리기술을이용한 멀티칼러 전송기법을 제안한다. 멀티칼라 부호화 및 이미지 처리기반의 복호화 기법을 제안한다.

This paper introduces the Optical Camera Communications (OCC) using image processing technique.The architecture andoperation of OCC system are given. To enhance data rate which is limited by sampling operation of commercial 30fpscamera, multi colors transmission technique is employed, leading to the importance of color image processing technique. Multi color encoding and image processing based decoding will be proposed in the paper.

본 연구에서는 저에너지(80kVp) 및 고에너지(140kVp) 이중 에너지 CT(DECT)를 이용하여 획득한 척추 영상에서 통풍 병변을 검출하고 시각화하는 영상 처리 기법을 제안한다. 회색조 동시 발생 행렬(GLCM) 분석 결과, 두 에너지 레벨 간에 명암 대비 및 에너지 값에 유의미한 차이가 관찰되었지만, 균질성 및 상관관계는 상대적으로 변화가 없었다. 이산 웨이블릿 변환(DWT) 분석에서는 80kVp 영상에서는 유사한 픽셀 강도가 넓게 분포하는 반면, 140kVp 영상에서는 특히 뼈 구조에서 고강도 영역이 집중되는 것을 확인하였다. 본 연구 결과를 검증하기 위해 Levenberg-Marquardt 알고리즘을 이용한 신경망을 학습시 켰다. DWT에서 추출한 고주파 성분을 입력 특징으로, 두 에너지 레벨의 강도값을 출력 목표로 설정하였다. 적합 결과는 예측값과 실제값 사이의 높은 일치도를 보여주어 모델의 정확성을 입증했다. 이 방법은 통풍 병변을 효과적으로 식별하고 요산나트륨(MSU) 결정 침착을 명확하게 시각화하는 것을 가능하게 하였다.본 연구는 기존의 이중 에너지 컴퓨터 단층촬영(DECT)을 이용하여 병변의 상세한 위치 파악과 물질 감별을 가능하게 함으로써 척추 통풍 진단의 정확도를 향상시키는 데 중요한 기여를 할 것으로 생각한다. 또한, 치료 반응을 모니터링하고 치료 효과를 평가하는 데 유용한 도구로 활용될 수 있을 것이다.

In this study, I propose an image processing approach for detecting and visualizing gout lesions in spinal images obtained using dual-energy computed tomography (DECT) at low (80 kVp) and high (140 kVp) energy levels. Through gray-level co-occurrence matrix (GLCM) analysis, I observed significant differences in contrast and energy values between the two energy levels, while homogeneity and correlation remained relatively unchanged. Discrete Wavelet Transform (DWT) analysis further revealed that the 80 kVp images displayed a broad distribution of similar pixel intensities, whereas the 140 kVp images exhibited concentrated high-intensity regions, particularly in bony structures. To validate my findings, I trained a neural network using the Levenberg– Marquardt algorithm. I used the high-frequency components extracted from DWT as input features and the intensity values from both energy levels as target outputs. The fitting results showed a strong alignment between the predicted and actual values, confirming the accuracy of the model. The method effectively identified gout lesions and clearly visualized monosodium urate (MSU) crystal deposition. I believe this approach offers a valuable contribution to improving diagnostic accuracy in spinal gout by enabling detailed lesion localization and material differentiation using conventional DECT. Furthermore, it may serve as a useful tool for monitoring therapeutic response and evaluating treatment efficacy.

Today, there is a developing requirement for observation so as to keep up the dignity at a spot and guarantee the wellbeing and security of its kin. An airborne observation framework will be advantageous in such a manner. This paper depicts how an ethereal reconnaissance framework can be assembled utilizing an unmanned flying vehicle or an automaton. We start by depicting the highlights of our ethereal observation framework and afterward talk about a portion of the advances that we have utilized in building it. From that point onward, we notice how we have consolidated those innovations into an automaton and have made them cooperate amicably so as to accomplish our goal. This goal will be achieved with the help of cloud services and cellular service using 5G technology. It tends to be utilized in peacekeeping exercises and furthermore constant observing of a spot whenever of the day. The point is to give quick and productive reconnaissance at a moderate rate with the goal that it tends to be utilized broadly at the private, institutional and administrative levels.

최근에 정밀 추적이나 타격 등의 군사 목적으로 UAV를 활용하는 연구가 매우 활발하게 진행되고 있다. 앞서가는 비행체를 추적하기 위해 후방에서 촬영한 영상을 활용하여 롤, 피치, 요와 같은 그 비행체의 자세 정보를 추정하는 기존의 연구가 진행되었다. 본 연구에서는 기존의 영상처리기법을 이용한 연구에 칼만 필터를 적용함으로써 자세 정보를 더욱 정 밀하게 추정하는 방법을 제시한다. 영상처리를 사용해서 추정한 비행 자세 데이터에 칼만 필터를 적용함으로써 기존의 방 식에서 발생했던 자세 각도의 추정오차 범위를 크게 줄일 수 있었다. 시뮬레이션 실험을 통해서, 칼만 필터를 적용할 경우 비행체의 자세 정보를 더욱 정확하게 추정할 수 있음을 확인할 수 있었다.

In recent years, researches utilizing UAV for military purposes such as precision tracking and batting have been actively conducted. In order to track the preceding flight, there has been a previous research on estimating the attitude information of the flight such as roll, pitch, and yaw using images taken from the rear UAV. In this study, we propose a method to estimate the attitude information more precisely by applying the Kalman filter to the existing image processing technique. By applying the Kalman filter to the estimated attitude data using image processing, we could reduce the estimation error of the attitude angle significantly. Through the simulation experiments, it was confirmed that the estimation using the Kalman filter can estimate the posture information of the aircraft more accurately.