Earticle

In this study, we propose a deep learning-based forest road recognition technology and integrate it into an autonomous steering and speed control system for a forestry forwarder. We trained a YOLOv8 segmentation model on 640×480 pixel images collected from straight, curved, and occluded sections of forestry roads, achieving a precision of 0.997, a recall of 0.994, an F1 score of 0.995, and an average precision of 0.994. We developed a centerline extraction algorithm that calculates the lateral offset between the vehicle center and the detected road centerline using the segmented mask. This offset is then input into an electronic steering lever control system, which replaces the existing mechanical steering lever to enable rapid response. In field experiments conducted on actual forest roads, the maximum average path tracking error was 0.534 m over three trials, confirming that the vision-based approach can reliably guide the forwarder under various lighting conditions, shadows, and roadside structures. This method is expected to overcome the limitations of GNSS/INS-based autonomous driving systems, as it operates solely with cameras and single-board hardware. The research results suggest that deep learning-based image processing provides a solid foundation for fully autonomous forestry operations and has the potential to improve the productivity and safety of timber transport operations.

This research will make character image printed from stone. Currently from computer graphics field, the field which cartoon rendering is a emphasizes a variation and an outline about the image of existing and creates the image which is satirize. In image printed from stone, We respect the rehabilitation of the image which is damaged to detect an outline. and there is restriction where image oneself is black-and-white mode. In order to solve this problem, first will make image processing printed image of 12 animals relief, fairy relief, stone lamp relief. And, the image which is processed colour controlled .From this processing, These past stone relief currently lived and animated with the new character resurrected beautifully with art.

오늘날 자동차 기술의 핵심은 IT 기반 융합 시스템기술로 변화하고 있다. 다양한 IT 기술을 접목하여 운전 중 다양 한 상황에 대응하고 또한 운전자의 편의성을 지원하는 기술적 추세를 보이고 있다. 본 논문에서는 운전자의 안전성과 편의성을 증대하기 위해 영상 정보를 기반으로 도로 정보를 검출해 운전자에게 알려주고, 버튼을 직접 손으로 눌러야 하는 물리적 인터페이스를 대체할 비접촉식 인터페이스 기술을 융합한 Augmented Driving System (ADS) 기술을 제 안한다. 본 기술은 카메라로부터 입력 받은 영상 정보를 제안된 알고리즘을 통해 앞차와의 거리, 차선, 교통 표지판을 검출하고 차량 내부를 주시하는 카메라와 운전자의 음성을 인식할 마이크를 기반으로 기본 음성인식과 동작인식이 융 합된 인터페이스 기술을 제공한다. 이러한 요소 기술들은 운전자가 인지하지 못하더라도 운전자에게 현재의 주행상황 을 인지하여 자동으로 알려줌으로써 교통사고 확률을 크게 낮출 수 있을 것이며, 또한 다양한 운전 중 기능 조작을 편 리하게 지원함으로써 운전자의 전방 주시에 도움을 줄 수 있다. 본 논문에서 개발된 기술을 통해 테스트를 실시해 본 결과 표지판인식, 차선검출, 앞차와의 거리 검출 등의 인식률이 약 90% 이상이 되었다.

Today, the core technology of an automobile is becoming to IT-based convergence system technology. To cope w ith many kinds of situations and provide the convenience for drivers, various IT technologies are being integrated int o automobile system. In this paper, we propose an convergence system, which is called Augmented Driving System (ADS), to provide high safety and convenience of drivers based on image information processing. From imaging sen sor, the image data is acquisited and processed to give distance from the front car, lane, and traffic sign panel by the proposed methods. Also, a converged interface technology with camera for gesture recognition and microphone fo r speech recognition is provided. Based on this kind of system technology, car accident will be decreased although d rivers could not recognize the dangerous situations, since the system can recognize situation or user context to give attention to the front view. Through the experiments, the proposed methods achieved over 90% of recognition in ter ms of traffic sign detection, lane detection, and distance measure from the front car.

The Nepalese Currency Counterfeit Detection System aims to address the growing threat of counterfeit currency in Nepal. Counterfeit banknotes pose a significant risk to the nation's financial stability and erode public trust in the monetary system. This system seeks to differentiate genuine banknotes from forgeries, thereby ensuring the integrity of financial transactions. Counterfeiting activities have a detrimental impact on the livelihoods of individuals and the overall economy. To combat this issue, this project will thoroughly investigate the various security features embedded within Nepali currency. Subsequently, it will leverage advanced image processing and computer vision techniques to develop a software-based solution capable of detecting and validating the authenticity of Nepali banknotes. While specialized machines are available in banks and commercial establishments for currency authentication, these systems are often inaccessible to the general public. This project aims to bridge this gap by developing a user-friendly software application that empowers individuals to independently verify the authenticity of Nepali currency.

본 연구는 스마트팜 자동화를 위한 딥러닝 기반 엽채류 이미지 분류 시스템을 개발하고, 그 성능을 심 도 있게 분석하였다. 기본적인 3층 구조의 CNN 모델뿐만 아니라 ResNet, Inception과 같은 고도화된 딥러닝 모델을 활용하여 엽채류 이미지를 분류하였으며, 각 모델의 성능을 비교하였다. 연구에 사용된 데이터는 스마 트팜 환경에서 촬영된 고해상도 이미지와 공개 데이터셋으로 구성되었으며, 다양한 전처리 및 데이터 증강 기 법을 통해 모델의 학습 효과를 극대화하였다. 그 결과, 딥러닝 기반 분류 시스템은 기존의 원본을 기반으로한 딥러닝에 비해 더 높은 정확도와 효율성을 보여주었으며, 스마트팜의 자동화를 통한 생산성 향상과 비용 절감 에 기여할 수 있음을 확인하였다. 이러한 연구는 스마트팜 기술의 발전에 중요한 역할을 할 것이며, 농업 자동 화의 미래를 위한 기술적 토대를 마련하는 데 기여할 것으로 기대된다.

This study developed a deep learning-based leafy vegetable image classification system aimed at smart farm automation and conducted an in-depth performance analysis. In addition to the basic 3-layer CNN model, advanced deep learning models such as ResNet and Inception were employed to classify leafy vegetable images, and the performance of each model was compared. The dataset used for the study consisted of high-resolution images captured in a smart farm environment, along with publicly available datasets. Various preprocessing and data augmentation techniques were applied to maximize the effectiveness of model training. As a result, the deep learning-based classification system demonstrated higher accuracy and efficiency compared to traditional deep learning approaches based on original data. It was confirmed that the system could contribute to improving productivity and reducing costs through smart farm automation. This research is expected to play a key role in advancing smart farm technologies and contribute to laying the technological foundation for the future of agricultural automation.

기존의 실내위치 인식 기술은 WLAN, GPS, Bluetooth, 또는 자기장 기술을 사용하는 것이 일반적이다. 하지만 이러한 기술은 실내 환경에서는 수신 신호 세기로 인해 오차를 발생시킬 수 있다. 센서를 사용하지 않는 실내위치 인식 방법으로 3D 건물 모델을 참조하는 방법이 있다. 본 논문은 3D 건물 모델을 참조하기 위하여 설계도면을 추 정 하는 방법을 제안한다. 제안된 방법은 이미지를 사용하여 이미지 안의 문과 같은 건물 안의 영구적인 정적 물체 를 인식, 사용하여 정적 물체와 정적 물체 간의 실제 거리 비율을 예측, 계산한다. 계산한 정적 물체 간의 실제 거리 비율을 사용하여 이미지 안의 정적 물체를 재배열하여 설계도면을 모델링한다. 본 논문의 설계도면 모델링 방법을 사용하여 실내위치 인식을 하면 특별한 인프라나 센서 없이도 이미지 한 장으로 값싸게 실내위치 인식을 할 수 있을 것으로 기대된다.

In the conventional indoor location recognition technology, WLAN, GPS, and Bluetooth technologies are generally used. However, this technique may cause an error in the indoor environment due to the received signal strength. There is a method of reference from 3D building model as a method of recognizing indoor location without a sensor. This paper proposes a method for modeling blueprint to refer to a 3D building model, which is one of the methods of indoor location recognition. The proposed method uses images to recognize static objects in buildings, such as doors in images, to predict and calculate the real-world distance ratio between static objects and static objects. Then our system model the blueprint by rearranging the static objects in the image using the calculated ratio of actual distances between static objects. It is expected that the indoor location recognition using the design drawing modeling method of this paper will be able to recognize the indoor location inexpensively with a single image without special infrastructure or sensor.

In this work, we implemented a fatigue detection system using machine learning and evaluated its performance. The proposed system is mainly based on the Viola-Jones face detection algorithm and the convolutional neural network (CNN). Viola-Jones object detection framework is mainly focused on the detection of the face and facial features. CNN is extended to DenseNets and made fully convolution to tackle the problem semantic image segmentation. The main idea behind the DenseNets is to capture the dense blocks that perform iterative concatenation of feature maps. The proposed system is implemented on many different video sequences and observed that its average accuracy is 99.18% and the detection rate is 99.71% out of approximately 100 image frames. The system shows high accuracy in segmentation, low error rate, and quick processing of input data distinguishes from the existing similar systems. Finally, if this system is implemented, it can minimize the number of accidents caused by drivers' fatigue.

목적 : 본 연구는 비장애 아동을 대상으로 한글쓰기 숙련도가 시선고정에 반영되는지를 알아보고자 하였다. 연구방법 : 만 7-9세 비장애 아동 21명이 본 연구에 참여하였다. 아동들은 Tobii Pro Glasses 2를 착용하고 10개 단어로 구성된 한글쓰기 과제를 수행하였다. 글자 크기 변이계수로 글씨쓰기 숙련도를 측정하였고, 시선추적장치에서 기록된 시선 위치 자료로부터 시선고정 빈도를 추출하였다. 글자 크기 변이계수와 시선고정 빈도 간의 상관분석을 실시하였다. 연구결과 : 세로크기 변이계수는 초당 시선고정 빈도와 정적으로 유의미한 상관을 나타냈다 (r = .459, p < .05). 이는, 글자 세로크기를 일관되게 쓰는 아동은 시선을 적은 위치에 고정함을 의미한다. 결론 : 본 연구는 비장애 아동을 대상으로 한글쓰기 수행과 시선고정 간의 관련성을 제시하였다. 본 연구는 한글쓰기 수행 동안 시선 움직임을 측정한 첫 시도로 향후 한글쓰기 발달 연구의 방향을 제시할 수 있을 것이다.

Objective : This paper aimed to measure the relationship between the performance of Korean handwriting and visual fixation for children. Methods : Twenty-one typically developing children aged 7 to 9 years participated in the study. The children performed Korean handwriting task wearing Tobii Pro Glasses 2. The Korean handwriting task consisted of 10 words from elementary school textbooks. The handwriting skill was measured by the coefficient variation of the letter size and the fixation cound and duration. Correlation analysis was performed to investigate the relation between visual fixation and the coefficient variation of the letter size. Results : The results showed that the visual fixation per second was positively correlated with Korean handwriting vertical size coefficient variation, indicating that the more consistent the vertical size of the letter, the smaller the fixation count per second. Conclusion : The results suggested a relation between the performance of Korean handwriting and visual fixation in typically developing children. This study is the first attempt to measure eye movement during the Korean handwriting process, and suggests a future direction for research on students’ development in writing Korean.

본 논문에서는 직선을 판별하는 판별방법과 또한 직선으로 판정되는 경우 직선의 8가지 방향에 대한 판정 을 위하여 좌표정보와 삼각함수 성질을 이용하여 직선 8반향 알고리즘을 제안하였다. 본 실험은 각 동작 별로 100회씩 총 800회를 실험 하였으며, 본 논문에서 제안한 알고리즘을 통한 정확도는 왼쪽 위 로의 대각선 방향이 92%로 가장 높았으며 왼쪽 방향, 오른쪽 위로의 대각선 방향과 오른쪽 아래로의 대각선 방향이 모두 82%로 가장 낮은 인식률을 나타내었다. 이 방법은 기존 방법과는 다르게 학습과정 없이 이미지 처리를 통해 얻은 좌표 정보만으로 손 제스처 인식이 가능 하다는 것을 보여준다.

In this paper, we proposed the straight line determination method and the algorithm for 8 directions determination of straight line using the coordinate information and the property of trigonometric function. We conduct an experiment that is 8 hand gestures are carried out 100 times each, a total of 800 times. And the accuracy for the 8 derection determination algorithm is showed the diagonal direction to the left upper side shows the highest accuracy as 92%, and the direction to the left side, the diagonal direction to the right upper side and the diagonal direction to the right bottom side show the lowest accuracy as 82%. This method with coordinate information through image processing than the existing recognizer and the recognition through learning process is possible using a hand gesture recognition gesture.

최근 미국에서 TV를 제어할 때 기존의 리모컨 대신에 손동작을 이용하는 기술이 상용화되었다. 하지만 그것은 리모컨의 역할을 사람의 손동작으로 TV가 인식하는 응용기술로 써 실질적인 에너지 절약과는 거리가 있었다. 본 논문에서는 TV가 사용자의 모습을 영상처리를 하여 실제로 시청을 하지 않는 경우 전원을 제어하며 에너지를 절약하고 시청하지 못한 컨텐츠는 자동 저장하여 사용자가 원하는 시간에 다시 볼 수 있게 하는 기술을 제안한다. TV에 설치된 카메라 모듈을 통해 주기적으로 사용자의 모습을 촬영하고 Gabor 필터를 이용한 영상 처리로 사용자의 상태를 감시하여 잠이 들었다고 판단될 경우 TV 전원을 자동으로 차단시킨다. 스마트 미터를 도입한 실험 결과를 통해 제안 시스템을 사용하게 되면 기존대비 약 4.8배의 전력 소모를 줄일 수 있음을 확인할 수 있다.

Recently, the technology that controls televisions through users' hand movements instead of remote controls has been commercialized in the United States. However, it is only the substitutional applied technology of remote controls that recognizes a certain set of functions, so that this technology does not effectively economize the energy used by the equipment. In this paper, this new technology will be proposed. The technology provides a solution to prevent wasting power by image processing when the television is not being used. In addition, the new technology is able to record programming contents while the television is off, thus the user is capable of possessing a choice of which time the contents are being watched. Moreover, the motions of the user are continuously captured by the camera module installed in the television. The image data processed through image processing such as gabor filter delivers the method to monitor the state of the user. If the user falls asleep, the television shut itself down to save the power. Experimental results of the introduction of smart meters through the proposed system, if you use about 4.8 times compared to existing models that can reduce power consumption can be found.

이 논문에서는 추사 김정희에 대한 콘텐츠 작업을 통해 그것을 실용적으로 활용할 수 있는 방법을 모색하였다. 먼저 오늘날 소비자 입장에서 추사 글씨에 대한 미학적 특성으로 복합적 미감에 주목하였다. 그의 글씨는 고졸하면서 웅혼하고 기괴하면서도 허화의 경지가 있다. 유가적 선비 기풍과 불가의 선적 취향이 함께 느껴지고 단조로운 것 같으면서도 복합적이다. 사람들이 그의 글씨에 매료하는 까닭은 바로 그의 글씨가 던지는 복합적 아름다움에 있다고 보았다. 추사 글씨를 콘텐츠화하면 여러 분야에서 활용할 수 있다. 먼저 서체를 필요로 하는 비문이나 현판, 명함으로 사용할 수 있다. 추사 서체의 시스템 개발로 추사체 폰트를 개발할 수도 있다. 디자인 측면에서는 타이포그래피와 같은 디지털 아트로 활용할 수도 있다. 교육적인 측면에서 학습 콘텐츠로 사용할 수도 있고 학술적으로도 활용할 수 있다. 그것에 앞서 무엇보다도 추사 글씨의 원천 자료를 확정하여 진위 여부를 확인해야 한다. 그리고 이미지 프로세싱, 데이터베이스 설계 과정과 같은 추사 서체의 디지털화를 거친다. 1) 추사 서체를 디지털화를 위해서는 가장 먼저 진본의 추사 서체를 스캔한다. 2) 그것을 일정한 크기로 잘라내어 스캔한 이미지 중 글자의 부분을 추출한다. 3) 데이터베이스를 설계하여 추출된 추사 서체의 개별 이미지들을 기존의 한자사전 데이터베이스를 이용하여 적용할 수 있도록 한다. 한 마디로 추사에 대한 문화콘텐츠화는 그를 과거의 추사에서 현재 살아 있는 추사로 재생시킨다. 이를 바탕으로 향후 여러 명필에 대한 서체 콘텐츠 개발에 응용하여 문화적으로나 학술적으로 보탬이 될 수 있겠다.

In this study, through contents work on Chusa Kim Jeong-Hee's style of handwriting, methods to practically utilize it was sought for. First, as an aesthetic characteristic of Chusa style of handwriting, the complex sense of beauty was noted. If Chusa style of handwriting is made into contents, the contents can be utilized in many areas. The diverse Chusa fonts can be developed into a system or new Chusa fonts can be developed. In the aspect of designs, they can be utilized as digital arts such as typography. In the aspect of education, they can be use as study contents or can be utilized scientifically. Digitizing of Chusa fonts undergoes image processing and database design processes. To digitize Chusa fonts, the original Chusa fonts should be scanned first. Then, the scanned images should be cut to pieces in a certain size to extract the part of letters from the scanned images. Then, a database should be designed so that the individual images of Chusa fonts extracted as such can be applied using existing Chinese character dictionary databases. The database can be applied to the development of font contents from many excellent handwriting styles so that the database can be helpful culturally or scientifically.

본 논문에서는 모서리 검출을 이용하여 추출된 모서리를 추적함으로써 차선을 검출하고 추적하는 알고리즘을 제안한다. 제안하는 알고리즘은 양 차선이 연속되지 않고 끊긴 형태로 존재하거나 교차되는 등의 다양한 차선의 형태에도 높은 추출률을 보이는 장점을 가지고 있다. 이는 이러한 형태의 차선의 비율이 높은 시내도로 와 국도에서의 차선 추출에 보다 유리하다. 이러한 점을 증명하기 위해 테스트는 주로 불연속적이고 교차되는 형태의 차선이 많은 도로에서 실시하였고 평균 87% 이상의 추출률을 보여주었다.

This paper presents an algorithm for tracking lanes on the road based on corner detection techniques. The proposed algorithm shows high accuracy regardless of lane divider types, eg, solid line, dashed line, etc, and thus is of advantage to city streets and local roads where various types of lane dividers are used. A set of experiments was conducted on real roads with various types of lane dividers and results show an extract ratio over 87% in average.

Bones are dynamic structures, being supported by muscles, tendons, and ligaments. When some or all the structures are disturbed i.e. in fractures, the alignment of the bone in respect to the rest of the body is deranged. This gives rise to axial as well as rotational deformity in three dimensional planes. The correct alignment and position of the long bones are to be maintained to heal the bone in the best possible anatomical and functional position. The objective of this research is to address the problems in the current practice involving surgeon, assistant, fluoroscopy and crude mechanical means and to see if a robotic solution exists to solve the problems of manipulating and reducing long bone fractures. This paper presents various design aspects of the proposed surgeon-instructed, image-guided and robotic system including the system design specification, robot design and analysis, motion control and implementation, and x-ray image processing and incorporation in CAD environment.

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.

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.