Machine vision is a new technology developed in recent years, it is the use of optical mechatronics means the machine has a visual function. The introduction of machine vision into the field of detection, in many occasions to achieve high-precision online high-speed measurement. At the same time, the theory of machine vision detection technology also grows step by step, can be divided into the following several development course and development trend. 1. Primary Vision Theory: The inverse problem that focuses primarily on optical imaging consists of a series of processes that restore the physical properties of a three-dimensional visible surface from a two-dimensional light intensity array. Here the process of input data and calculation purposes are able to clearly describe, such as edge detection, stereo matching, by the motion recovery structure and other methods. In the process of projecting a three-dimensional object into a two-dimensional image, the three-dimensional information has a lot of losses, which leads to the ill-posed problems. Therefore, it is particularly important to study the primary vision process and its constraints, which is mainly aimed at 3D reconstruction and binocular vision. 2. Active Vision Theory: Active vision refers to the observer to determine or indefinite way to track the target movement, perception of the object of technical methods. In active vision, observers and target objects can also move simultaneously. Observer movement provides additional conditions for studying the shape, distance and movement of the target. Important research directions are target tracking and missile interception. 3. Visual information fusion: the fusion of multiple visual information, may break the limitations of access to a single visual information to achieve the use of the ideal environment static and transient visual access to information, to meet the complex objective world requirements, the main research areas Image information fusion. 4. Reconstruction of 3D scene: At present, the theory and algorithm of 3D scene restoration are limited to the "visible" part of the scene. It belongs to the 2.5-dimensional information expression and only provides the 3D information within the visible outline of the object. Restoring the complete information of the visible and invisible parts of the surface of a scene is a complicated but urgent theoretical problem to be solved. 5. Algorithmic performance evaluation: Machine vision research concerns whether the task can be completed or not, the lack of performance evaluation algorithms and systems approach quality. In practical applications, efficiency and performance are very important, otherwise the algorithm and system can not get out of the laboratory. Therefore, the establishment of the performance evaluation of the machine vision algorithm is indispensable. 6. Visual Parallel Computing: Visual Real-Time Computing There are many theoretical, algorithmic and technical issues. The trend of visual parallel computing architecture is to adopt smaller and smaller processing units in larger and larger structures. The development direction of the visual parallel computing structures is the processing network system composed of basic logic operation processing units. 7. Universal Visual Information System: a universal visual information system that can accomplish various visual tasks, namely, to establish a machine vision system that is analogous to the function of the human visual system. By establishing a dedicated visual system platform, it gradually develops into a complete universal visual system such as visual Platform, a highly intelligent visual robot. Laser Welding Machines,PortableLaser Welding Machine,Welding Machine Laser,Laser Welding Machine 2000W JIANGSU FUERMU WELDING CORPORATION , https://www.fuermuwelding.com