China 
Africa 
Explore chapters and articles related to this topic
Computer and Human Vision Systems
Published in Sheila Anand, L. Priya, A Guide for Machine Vision in Quality Control, 2019
Computer and machine vision engage techniques of image processing to understand the image or objects being viewed. Image processing techniques, in general, have an image as input and give a modified image as output. The images may be filtered, smoothed, sharpened, or even converted from color to gray scale. Computer vision and machine vision both use image processing to understand images but this is followed by interpretation to arrive at a decision or logical conclusion as output. In the case of machine vision, this decision or logical conclusion is often followed by some autonomous action in the industry.
Robotics and Machine Learning
Published in Shivani Agarwal, Sandhya Makkar, Duc-Tan Tran, Privacy Vulnerabilities and Data Security Challenges in the IoT, 2020
Machine vision or robot vision, also known as computer vision, is based upon computer algorithms, as robotics experts use a camera for robotic vision to capture physical data. The robot vision is coupled with machine vision, assisting the former to boost the robot-assisted human guidance and machine prediction. The small differences between the two terms may be ignored as robotics involves the adjustment of the frame of reference and the capacity of the robot to physically alter its surroundings.
Applied Machine Vision in Agriculture
Published in Guangnan Chen, Advances in Agricultural Machinery and Technologies, 2018
Machine vision systems are comprised of camera and computer hardware, software algorithms, and memory that interact in a conceptually similar manner to our human visual system to extract information from images and provide decision support or automation within a system. The following section presents an overview of the common processes, hardware and software technologies, and machine learning and image analysis techniques that are the foundation of machine vision systems.
An overview of current technologies and emerging trends in factory automation
Published in International Journal of Production Research, 2019
Mariagrazia Dotoli, Alexander Fay, Marek Miśkowicz, Carla Seatzu
Machine vision is a technical area that requires extended computational resources and therefore feels a problem of fundamental limits of conventional computing in terms of processing speed, power consumption, reliability and programmability (The Human Brain Project 2012). A promising direction for future vision technology development is the adoption of bio-inspired event-based dynamic vision sensors which asynchronously detect the significant relative light intensity changes in a scene and output them in a form of address-event representation. Instead of sending entire images at fixed frame rates, Dynamic Vision Sensors (DVS) report only local pixel-level changes caused by movement in a scene which allows a huge reduction of power consumption, data storage and computational requirements of the vision system. Due to high energy efficiency, the DVSs are attractive for applications of mobile robots. In general, the DVSs outperform the traditional imaging especially for passive vision systems (i.e. machine vision systems operating on static images). The problem of 3D image reconstruction for DVSs is based on stereo matching of pseudo-frames produced from DVS events accumulated over a particular period of time. Existing methods for asynchronous stereo vision with DVSs are extended by using the cooperative approach, in which the history of the recent activity in the scene is stored to serve as spatiotemporal context used in disparity calculation for each incoming event.
A real-time intelligent classification model using machine learning for tunnel surrounding rock and its application
Published in Georisk: Assessment and Management of Risk for Engineered Systems and Geohazards, 2023
Junjie Ma, Tianbin Li, Gang Yang, Kunkun Dai, Chunchi Ma, Hao Tang, Gangwei Wang, Jianfeng Wang, Bo Xiao, Lubo Meng
Convolutional Neural Network (CNN) is one of the representative algorithms of deep learning, which was first reported in 1989 for machine vision (LeCun et al. 1989). In the past 20 years, a series of classification networks with different architectures have been proposed based on the CNN principle, such as LeNet, AlexNet, GoogleNet, VGGNet, ResNet, etc (Khan et al. 2020). In particular, VGGNet and ResNet have significantly improved network depth, computing speed, and complexity. As a result, VGGNet and ResNet have been widely used in medicine (Younis 2021; Liu, Zhou, and Peng 2022), industry (Wen, Li, and Gao 2020; Zhang, Lv, and Cheng 2020), civil engineering (Samma et al. 2021; Chou et al. 2022), and agriculture (Saleem, Potgieter, and Arif 2021; Tao and Wei 2022).
Digital technologies for energy efficiency and decarbonization in mining
Published in CIM Journal, 2023
Machine vision (also known as computer vision) uses technology to mimic human visual capabilities, integrating image processing and analysis techniques for automation and optimization purposes (Lange, 1992). Cameras installed on mining equipment and coupled with algorithms and drones equipped with cameras and sensors have the potential to improve the safety and efficiency of various stages of the mining cycle.