• <s id="q8qog"></s>
  • <option id="q8qog"></option>
  • <td id="q8qog"><blockquote id="q8qog"></blockquote></td>
  • <noscript id="q8qog"></noscript>
  • 迪奧科技標語
    Problems with machine vision technology
    Author:Administrator   Published in:2020-01-02 15:15

    Machine vision mainly studies the use of computers to simulate human visual functions. Information is extracted from images of objective things, processed and understood, and ultimately used for actual detection, measurement, and control.

    In recent decades, the vision system has been widely used in agriculture, industry, medicine and other fields due to its outstanding advantages such as non-contact, high speed, high precision, and strong anti-interference ability on the spot. As long as it is necessary to recognize, judge and detect objects, machine vision can flex its muscles and complete tasks quickly and well. However, we have to admit that machine vision still has shortcomings in practical applications. For the human eye, it may be easy to recognize an object and understand the surrounding scene, but for machine vision, there are several problems:

    1. Multiple meanings in an image

    When a three-dimensional image is projected into a two-dimensional image, a lot of information is lost. Perhaps two identical two-dimensional images are projected from two completely different three-dimensional scenes. In addition, the same object, taken from different visual angles, will also produce different images, and it will be very difficult for the machine to distinguish.

    2. Influence of environmental factors

    The surrounding noise, the instability of light, the shape and color of the object, and the changes in the mutual position of the camera and the object will all affect the image to a certain extent.

    3. Large amount of data

    Whether it is a color image or a gray image, the amount of information is very large, so it requires a large storage space, so it is not easy to achieve fast processing.

    4. Real-time issues

    Real-time is mainly divided into hard real-time and soft real-time. With the development of hardware, there have been better solutions for hard real-time problems.

    The soft real-time problem is mainly solved by algorithms to improve the processing speed of the image. But in this case, there will be problems of accuracy and robustness reduction in complex work sites. Therefore, it is impossible to accurately obtain an image of an object.

    After decades of development, machine vision has matured. This greatly improved the level of machine automation and intelligence, and brought huge economic benefits to the society. Although there are still some difficulties in solving problems such as image resolution and real-time performance, with the development of science and technology, these problems will also be well resolved. Machine vision is also an emerging industry in China. More and more companies and research institutions have begun to study machine vision. Machine vision will inevitably become China's pillar industry.

    The above is an introduction to the problems of machine vision technology. Do you have a new understanding of machine vision? If you want to learn more about machine vision, welcome to follow TEO Technology.

    Copyright @ 2016 - 2019 迪奧科技 版權所有 備案號: