• 한국컴퓨터정보학회논문지
• /
• 제28권9호
• /
• pp.9-15
• /
• 2023

컴퓨터 비전 분류 연구에서 합성곱 신경망 (Convolutional Neural Network)은 탁월한 이미지 분류성능을 보여준다. 이에 영감을 받아 예술 관련 이미지 분류 작업에 대한 적용 가능성을 분석해 본다. 본 논문에서는 예술 작품 아티스트 분류의 정확도를 향상시키기 위해 최적화된 합성곱 신경망 구조를 제안한다. 미세 조정 범위 시나리오와 완전연결층 조정 시나리오를 세운 뒤 그에 따른 예술 작품 아티스트 분류의 정확도를 측정했다. 즉, 학습 컨볼루션 레이어(Convolution layer) 수와 완전연결층 수 등 ResNet50 모델의 구조를 변경하며 예술 작품 아티스트 분류의 정확도가 향상되도록 최적화했다. 본 논문에서 제안하는 합성곱 신경망 구조는 기존 예술 작품 아티스트 분류에서 쓰이던 AlexNet 모델을 1-GPU 버전으로 수정한 CaffeNet 모델보다 더 높은 정확도를 실험결과에서 증명한다.

• Journal of Animal Science and Technology
• /
• 제66권1호
• /
• pp.31-56
• /
• 2024

Pig farming, a vital industry, necessitates proactive measures for early disease detection and crush symptom monitoring to ensure optimum pig health and safety. This review explores advanced thermal sensing technologies and computer vision-based thermal imaging techniques employed for pig disease and piglet crush symptom monitoring on pig farms. Infrared thermography (IRT) is a non-invasive and efficient technology for measuring pig body temperature, providing advantages such as non-destructive, long-distance, and high-sensitivity measurements. Unlike traditional methods, IRT offers a quick and labor-saving approach to acquiring physiological data impacted by environmental temperature, crucial for understanding pig body physiology and metabolism. IRT aids in early disease detection, respiratory health monitoring, and evaluating vaccination effectiveness. Challenges include body surface emissivity variations affecting measurement accuracy. Thermal imaging and deep learning algorithms are used for pig behavior recognition, with the dorsal plane effective for stress detection. Remote health monitoring through thermal imaging, deep learning, and wearable devices facilitates non-invasive assessment of pig health, minimizing medication use. Integration of advanced sensors, thermal imaging, and deep learning shows potential for disease detection and improvement in pig farming, but challenges and ethical considerations must be addressed for successful implementation. This review summarizes the state-of-the-art technologies used in the pig farming industry, including computer vision algorithms such as object detection, image segmentation, and deep learning techniques. It also discusses the benefits and limitations of IRT technology, providing an overview of the current research field. This study provides valuable insights for researchers and farmers regarding IRT application in pig production, highlighting notable approaches and the latest research findings in this field.

• Journal of Information Processing Systems
• /
• 제13권5호
• /
• pp.1043-1051
• /
• 2017

• Journal of Information Processing Systems
• /
• 제13권6호
• /
• pp.1431-1435
• /
• 2017

• 전기전자학회논문지
• /
• 제23권1호
• /
• pp.249-253
• /
• 2019

현재 상용화된 악보 채보 프로그램은 오디오 정보를 기반으로 채보를 진행한다. 이러한 기존 채보 프로그램은 환경 의존성, 장비 의존성, 시간 지연이라는 단점을 지니고 있다. 본 논문은 기존의 오디오를 이용하여 채보를 방식을 지양하고, 연주 영상을 분석하여 채보를 진행하는 컴퓨터 비전 기반 악보 채보 시스템을 제안한다. 제안하는 악보 채보 시스템은 대중화된 스마트폰 카메라를 활용하여 피아노 연주를 촬영하고, 이를 분석하여 자동으로 전자 악보인 미디파일을 생성하는 방식으로 동작한다. 컴퓨터 실험에서 제안하는 악보 채보 시스템은 95.6%의 정확도로 연주된 음계를 채보하는 것으로 조사되었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제8권2호
• /
• pp.483-503
• /
• 2014

In this paper, human action recognition using pyramid histograms of oriented gradients and collaborative multi-task learning is proposed. First, we accumulate global activities and construct motion history image (MHI) for both RGB and depth channels respectively to encode the dynamics of one action in different modalities, and then different action descriptors are extracted from depth and RGB MHI to represent global textual and structural characteristics of these actions. Specially, average value in hierarchical block, GIST and pyramid histograms of oriented gradients descriptors are employed to represent human motion. To demonstrate the superiority of the proposed method, we evaluate them by KNN, SVM with linear and RBF kernels, SRC and CRC models on DHA dataset, the well-known dataset for human action recognition. Large scale experimental results show our descriptors are robust, stable and efficient, and outperform the state-of-the-art methods. In addition, we investigate the performance of our descriptors further by combining these descriptors on DHA dataset, and observe that the performances of combined descriptors are much better than just using only sole descriptor. With multimodal features, we also propose a collaborative multi-task learning method for model learning and inference based on transfer learning theory. The main contributions lie in four aspects: 1) the proposed encoding the scheme can filter the stationary part of human body and reduce noise interference; 2) different kind of features and models are assessed, and the neighbor gradients information and pyramid layers are very helpful for representing these actions; 3) The proposed model can fuse the features from different modalities regardless of the sensor types, the ranges of the value, and the dimensions of different features; 4) The latent common knowledge among different modalities can be discovered by transfer learning to boost the performance.

• 한국멀티미디어학회논문지
• /
• 제18권3호
• /
• pp.287-297
• /
• 2015

A railway security detection system is very important. There are many safety factors that directly affect the safe operation of trains. Security detection technology can be divided into passive and active approaches. In this paper, we will first survey the railway security systems and compare them. We will also propose a subway security detection system with computer vision technology, which can detect three kinds of problems: the spark problem, the obstacle problem, and the lost screw problem. The spark and obstacle detection methods are unique in our system. In our experiment using about 900 input test images, we obtained about a 99.8% performance in F- measure for the spark detection problem, and about 94.7% for the obstacle detection problem.

• ETRI Journal
• /
• 제27권5호
• /
• pp.539-544
• /
• 2005

An analog CMOS vision chip for edge detection with power consumption below 20mW was designed by adopting electronic switches. An electronic switch separates the edge detection circuit into two parts; one is a logarithmic compression photocircuit, the other is a signal processing circuit for edge detection. The electronic switch controls the connection between the two circuits. When the electronic switch is OFF, it can intercept the current flow through the signal processing circuit and restrict the magnitude of the current flow below several hundred nA. The estimated power consumption of the chip, with $128{\times}128$ pixels, was below 20mW. The vision chip was designed using $0.25{\mu}m$ 1-poly 5-metal standard full custom CMOS process technology.

• 반도체디스플레이기술학회지
• /
• 제9권3호
• /
• pp.59-63
• /
• 2010

This paper proposes computer visual inspection algorithms for various LCD defects which are found in a manufacturing process. Modular vision processing steps are required in order to detect different types of LCD defects. Those key modules include RGB filtering for pixel defects, gray-scale morphological processing and Hough transform for line defects, and adaptive threshold for spot defects. The proposed algorithms can give users detailed information on the type of defects in the LCD panel, the size of defect, and its location. The machine vision inspection system is implemented using C language in an embedded Linux system for a high-speed real-time image processing. Experiment results show that the proposed algorithms are quite successful.

• 한국전자통신학회논문지
• /
• 제16권2호
• /
• pp.211-218
• /
• 2021

건설 현장은 다양한 위험요소가 존재하고 있다. 안전재해를 줄이고자하는 다양한 접근이 있으나 어느 정도 한계성을 가지고 있다. IT의 무선통신 기술과 빠르게 발전하고 있는 이미지 처리 기술을 활용하여 위험요소를 사전에 식별하고 능동적으로 대응한다면 건설 현장에서의 재해를 감소시킬 수 있을 것이다. 따라서 본 연구에서는 건설 현장의 위험요소를 사전에 발견할 수 있는 시스템을 구성하고 실시간 컴퓨터 비전을 목적으로 한 OpenCV를 이용하여 건축현장의 위험요소를 발견하고 대응할 수 있도록 하는 시스템을 제안하였다.