Machine vision, also known as computer vision, uses computers to realize human visual functions, that is, using machines instead of human eyes to make measurements and judgments. Machine vision technology includes light source illumination technology, optical imaging technology, sensor technology, digital image processing technology, mechanical engineering technology, detection and control technology, analog and digital video technology, computer technology, human-machine interface technology and other related technologies. The basic technology.
The identification of traffic lights will make it possible for 7 to 8% of color-blind, weak-colored patients in the world to drive a car, and also advance the technology for driverless cars. Therefore, it will bring greater economic benefits and greater social benefits to the automotive industry and the automotive electronics industry, and can fill the gap in this field internationally.
2. Traffic light recognition method based on machine vision 2.1. Flow chart of traffic light recognition method is as follows
When acquiring an original image, we need to extract the part of the traffic light in the image, taking into account the changes in the background and the interference of other objects on the recognition of the traffic light. In this paper, the shape and gray value of the traffic light are used to locate the position of the traffic light in the image.
2.2.1, the squareness and circularity of the shape of the traffic light
The rectangular light of the traffic light can be used to find a certain range of the traffic light. Here, a simple rectangular degree calculation method is used, which is a discrete area of ​​the low gray value as the input area, when a certain rectangle is obtained. When the input area has the same first and second moments, the ratio of the area of ​​the input area to the area of ​​the rectangle is calculated, that is, the value of the rectangle degree. Obviously, when the input area is a rectangle, the maximum value of the rectangle is obtained by 1; the closer the input area is to the rectangle, the closer the rectangle is to 1 (the rectangle is 0 when there is no input area).
Through the above algorithm of rectangularity, it is possible to screen out a certain range (including the contour of the traffic light) in a region of low gray value, and finally locate the contour of the traffic light by the area occupied by the traffic light in the image.
Figure 2-1 Positioning the traffic light by shape
Then based on the above figure, a simple Circularity operator is used to determine the contour of the traffic light.
The specific algorithm is as follows:
Suppose F is the area of ​​a closed area, max is the maximum distance from the center point to the boundary or contour, then:
Circularity=F/(max^2*Ï€)(2-1)
According to Equation 2-1, the circularity of the circle is 1. It can be seen that for a region surrounded by a contour or a polygon, if its circularity is close to 1, then the contour approximates a circle. A contour similar to a circle can be selected by a threshold, for example, a contour with a circularity in the range of [0.8, 1] can be selected. If there are multiple contours, the areas corresponding to these contours are placed in an array.
2.3, color space changesAfter confirming the location of the traffic light, we need to determine the status of the traffic light by color recognition.
Since the similarity of the RGB color space cannot represent the similarity of colors, the HSI color space has no such problem. They are suitable for the human eye to distinguish and better reflect the perception and discrimination ability of people. So you can first convert the RGB color space into an HSI color space.
The general formula for converting RGB space into HSI space is as follows:
Figure 2-2 Traffic light information extracted by saturation
2.4, color recognitionThis paper uses image segmentation to identify the color of traffic lights. After the image is segmented by the selected threshold, find the desired image.
2.4.1. Threshold-based segmentation
This is one of the most commonly used area segmentation techniques, and the threshold is used to distinguish between different purposes.
The target gray value. If the image has only two categories, target and background, then just select a threshold called single threshold segmentation. In this method, the gray value of each pixel in the image is compared with a threshold value, pixels having a gray value greater than a threshold value are classified, and pixels having a gray value smaller than a threshold value are another type. If there are multiple targets in the image, multiple thresholds need to be selected to separate the targets. This method is called multi-threshold segmentation. The result of the threshold segmentation depends on the selection of the threshold, and the determination of the threshold is the key to the threshold segmentation. The threshold segmentation is essentially the process of finding the optimal threshold according to a certain standard.
Under the specific condition that the background and the prior probability of the target image are equal, the optimal threshold is the mean of the mean of the background gray and the mean of the gray of the target image. which is:
Waterproof RGB LED Dance acrylic Floor Display Screen for Wedding Party
These interactive display walls are applied in different purposes. This is a perfect audio/visual solution for commercial or
business purposes as well as for the educational purposes. It is often used for providing information to the viewers or audiences in the most interactive and effective way .
prospective customers. The interactive LED can be used IN retail stores, restaurants and other commercial spaces for giving information about their services and products.
1.Aluminum Structure Light Weight,good heat dissipation, good weather resistance, convenient transportation.
2.Imported PC Protrctive Surface, High Permeability, Strong Impact Resistance, Good anti-slip Effect
3.Strong Load Capacity The center of the panel has a load-bearing pivot,with a single panel bearing more then 1.2tons
Dance LED Display,Flexible Led Screen,Indoor Led Display Screen,Led Church Screen
Guangzhou Chengwen Photoelectric Technology co.,ltd , https://www.cwstagelight.com