Vision systems contribute to the industry’s main sales, and algorithms have entered the era of cloud intelligent computing

RDCU-12C Machine vision includes equipment manufacturing and system integration. Machine vision midstream enterprises must not only undertake upstream hardware equipment, but also carry out midstream algorithm development, but also provide overall solutions and related services for downstream customers. System integrators are mainly oriented to equipment manufacturers or end users, providing hardware integration, software services and other solutions, and providing non-standard services according to user requirements. At present, the threshold of the domestic system integration industry is low, the market competition is fierce, and the profit space of the industry is RDCU-12C relatively limited. Equipment manufacturers are not only doing software system integration, but also combining light sources, lenses, cameras, image acquisition cards, vision analysis software and software systems to develop for specific downstream application scenarios and manufacture complete machine vision systems or equipment.

RDCU-12C

RDCU-12C Vision systems and other equipment are the main components of machine vision industry sales. From the perspective of product composition, various types of machine vision products can be divided into three categories, the first category is a specific application vision system or configurable vision system, typical industrial machine vision systems include light source, lens, camera, image acquisition card, image processing software, communication, input and output units, etc., the second category is mainly hardware. Including cameras, optical components, lighting equipment, all kinds of sensors; The third category includes software, cables, acquisition cards and other accessories. According to the data of China Machine Vision Industry Alliance, in 2020, the sales of China’s feature application machine vision system accounted for 43.3%, 2D cameras for 18.8%, optical components for 16.2%, and the total proportion of the three reached 71.6%. Application vision system and hardware products accounted for nearly 90% in 2020, which is the main component of the machine vision market, RDCU-12C and the market share of software and accessories is small.

After four generations of innovation, machine vision algorithm has entered the era of cloud intelligent computing. Machine vision algorithm is crucial to the integration of midstream systems, and its development can be divided into four stages: the first generation is image processing for pattern recognition, which improves the quality of images; The second generation of pattern recognition image analysis, using feature extraction tools to realize the image target recognition, often principal component analysis, Fisher analysis, genetic algorithm, etc. The third generation image understanding deep learning can complete the classification, segmentation, detection and recognition functions; The fourth generation is cloud intelligent computing, which uses big data mining technology based on knowledge graph to establish a database of production experts. There are two main types of common machine vision algorithm tools, one is a toolkit containing a variety of processing algorithms, and the other is a special kind of application software to achieve a special kind of work, each of which has advantages and disadvantages, and needs to be selected according to the actual situation of the user.