Nowdays, based on OpenCV open source algorithm and deep machine learning AI technology, more and more traditional visual inspection jobs are replaced by "Visual Robot" or special "Computer vision software". As a part of patrol robot necessary technology, YZ Tech Co ltd also provide customized computer visual recognition service. Due to today's computer vision technology is still weak and stiff, so for each specifed application case, engineers have to develop a unique software to match this specific visual recognition task.
Here we list some of our field tested application samples for your ref.
When you want to learn more about visual recognition business, please contact us at once.

Meter Recognition system: For factory power devices room, one of the important tasks is monitoring the power devices data reading and the status of single lights in real time. Most meters and machines do not have automatic recording and remote alarm functions, so this "Meter Recognition system" is the best solution. The system is composed of PC computer and IPC webcam. They are connected by LAN wires. Each IPC camera aligns itself with the device meters in front, and sends the video information to the core computer through LAN to analyze and process.Our software use OpenCV algorithm and machine learning in AI technology to achieved this task.
Defective Shape Recognition: The right photo shows a real computer visual recognition application in a factory, which is manufacturing bottled drink. Using a fixed location camera, when a bottle come through the camera view channel, the camera take an image and feed it to our special designed software. The captured image is firstly prepared by gaussian filter and then convert from RGB color image to black-white image. Then use special contour detection algorithm, we can get the useful data informtion, then we can judge if this bottle covered good or not. We called this job is "shape defect recognition" or "detective shape recognition".
Defective Color Recognition: Color recognition is a basic function for computer vision algorithms. The right picture was taken from a packing production line field. In this application, the camera monitor each package box and see if all the required color items are correctly inside the package box. To support the camera, our engineer develop an compact algorithm based on OpenCV function library. Color-space conversion skill and photo pixel size value are the main factors of this software. One issue need be highlight: due to the HSV colors are very sensitive for different light source and different light project angle, so this system need a fixed light source and fixed camera aim angle.
Facial Recognition system: A facial recognition system is a technology capable of identifying or verifying a person from a digital image or a video frame from a video source. There are multiple methods in which facial recognition systems work, but in general, they work by comparing selected facial features from given image with faces within a database. YZ engineer using "CascadeClassifier" function from OpenCV to location all the faces in a frame and then use dlib-face recognition method to verify if the current face is in the registered face library. If your registered face library is less than 100 persons, our tiny off-line "YZ Face recognition system" will be a good selection for your application.
Object Recognition: Object recognition technology in the field of computer vision for finding and identifying objects in an image or video sequence. Humans recognize a multitude of objects in images with little effort, despite the fact that the image of the objects may vary somewhat in different view points, in many different sizes and scales or even when they are translated or rotated. This task is still a challenge for computer vision systems. Right photo shows a real application in factory. We developed a software to identify the valve switch status "OPEN" or "CLOSED". This is only a simple application. There are many different applications based on object recognition technology.