CN113240742A - Train reversing auxiliary detection method based on visual pixel link straight line detection - Google Patents

Abstract

The invention discloses a train reversing auxiliary detection method based on visual pixel link line detection, which comprises a line detection algorithm and a distance measurement algorithm, wherein the line detection algorithm is used for detecting a track based on a pixel link method, the principle of the pixel link algorithm is that the relation between each pixel on an image and the neighborhood of the surrounding pixels is judged, the core of the algorithm is to detect the similarity of each pixel of one picture, and if the pixel similarity threshold condition is set and meets the condition of the threshold value z, the algorithm can be used for detecting an iron rail line.

Description

Train reversing auxiliary detection method based on visual pixel link straight line detection

Technical Field

The invention relates to the technical field of visual detection, in particular to a train reversing auxiliary detection method based on visual pixel link straight line detection.

Background

Railways dominate the field of transportation in China, and bear a great mission to drive the development of national economy. The railway line in China has long distance and complex passing environment, foreign matter invasion is caused in front of a high-speed running train by random foreign matters such as landslides, detained people and the like caused by natural disasters, and the running safety is seriously influenced. Today, the demand of society for realizing automation cannot be met only by manual detection and fixed point installation of monitoring points in the rapid development of high-speed railways. With the realization of full-automatic driving in rail transit and the research and application of detection technology based on machine vision, how to realize intelligent detection of foreign matters in front of a high-speed train becomes a new trend of railway operation, and has irreplaceable practical significance. On the basis of the existing railway obstacle detection technology research at home and abroad, the invention realizes the obstacle detection by adopting a mode based on vehicle-mounted monocular machine vision. By analyzing the image of the acquired video frame, the segmentation algorithm of the railway track obstacle image is mainly researched, the algorithm is improved so as to achieve the purpose of improving the real-time property and the robustness, the improved algorithm is applied to a comprehensive judgment system to detect the obstacle, and whether the obstacle exists is judged according to two detection steps.

Disclosure of Invention

The invention aims to provide a train reversing auxiliary detection method based on visual pixel link line detection, which aims to solve the problems in the background technology.

In order to achieve the purpose, the invention provides the following technical scheme: a train reversing auxiliary detection method based on visual pixel link line detection comprises the following steps:

A. graying the original image;

B. carrying out denoising and filtering processing on the image of the gray level image, and processing the image polluted by noise by adopting a filtering method;

C. drawing an interest area in a picture to perform linear detection, so that a required linear line can be acquired more accurately;

D. detecting straight lines in an ROI area of the picture, filtering out horizontal straight lines and shorter straight lines, and taking only the longest line segment on each track;

E. extracting two detected paths and fitting the paths to an original image;

F. and detecting the distance between the obstacle and the lens.

Preferably, the image graying method in the step a is as follows:

a. respectively acquiring the energy of each color channel of the color image;

b. obtaining the optimal coefficient of each color channel according to the energy of each color channel;

c. and obtaining the gray level image corresponding to the color image according to the optimal coefficient of each color channel.

Preferably, an adaptive median filter is adopted in the step B to filter out noise.

Preferably, the specific method of step F is as follows:

a. calibrating and correcting the distortion of the camera;

b. deducing the corresponding relation between an image coordinate system and a physical coordinate system according to the imaging principle of the optical lens of the camera;

c. and shooting an image, and realizing accurate distance measurement between the train and the obstacle according to the relative position of the number on the image and the camera and the deflection angle.

Preferably, the distance measurement formula in step c is as follows:

preferably, α is a pitch angle, H is a camera height, θ is a vertical viewing field, a horizontal viewing angle is β, Dmin is an actual distance from a bottom edge of the image to the camera, Dmax is an actual distance from a top edge of the image to the camera, height is an image height, and coordinate points X and Y are a horizontal distance and a vertical distance from the camera.

Compared with the prior art, the invention has the beneficial effects that: the method can be used for detecting the iron rail line, the algorithm comprises image preprocessing, line detection, line fitting and distance prediction, and experiments show that the line detection based on the visual pixel linking method has wide applicability.

Drawings

FIG. 1 is a flow chart of the present invention;

FIG. 2 is a schematic diagram of an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Referring to fig. 1, the present invention provides a technical solution: a train reversing auxiliary detection method based on visual pixel link line detection comprises the following steps:

A. graying the original image;

B. carrying out denoising and filtering processing on the image of the gray level image, and processing the image polluted by noise by adopting a filtering method;

C. drawing an interested region in a picture for linear detection, so that the required linear can be more accurately acquired, and the problem of low processing speed can be effectively solved by only detecting the ROI region;

D. detecting straight lines in an ROI area of the picture, filtering out horizontal straight lines and shorter straight lines, and taking only the longest line segment on each track;

E. extracting two detected paths and fitting the paths to an original image;

F. and detecting the distance between the obstacle and the lens.

In the invention, the image graying method in the step A is as follows:

a. respectively acquiring the energy of each color channel of the color image;

b. obtaining the optimal coefficient of each color channel according to the energy of each color channel;

c. and obtaining the gray level image corresponding to the color image according to the optimal coefficient of each color channel.

In the step B, the image of the gray level image is subjected to denoising and filtering treatment, and the image polluted by noise can be treated by adopting a filtering method, but many linear filters have low flux, so that the edge is blurred while denoising, and the median filtering can remove the noise and protect the edge of the image under certain conditions, so that the method is a nonlinear noise removing method.

The median filtering is implemented by replacing the value of a point in the digital image by the median of the values of the points of a region of the point. We refer to a neighborhood of a certain length or shape of a point as a window, and for median filtering of two-dimensional images, a 3 x 3 or 5 x 5 window is typically used for filtering, but the speed of median filtering is slow.

The extreme value median filtering can know which points are noise points and which points are signal points, the noise points can be processed and the signal points are reserved, corresponding values are given to the noise points according to neighborhood correlation, such as neighborhood median, and the signal points are kept unchanged, so that image blurring can be reduced, the filtering process does not affect signals, and only the noise removing effect is achieved, and the method has the defect that the performance is greatly reduced when the noise is large.

In the case where the noise density is not so large, the effect of using median filtering is good. However, when the probability of noise occurrence is high, the original median filtering algorithm is not very effective. Only increasing the filter window size, although blurring the image. The purpose of using the adaptive median filter is to dynamically change the window size of the median filter according to preset conditions, so as to simultaneously take the effects of denoising and detail protection into consideration, and ensure real-time performance. And B, adopting adaptive median filtering to filter noise.

In the invention, the specific method of the step F is as follows:

a. calibrating and correcting the distortion of the camera;

b. deducing the corresponding relation between an image coordinate system and a physical coordinate system according to the imaging principle of the optical lens of the camera;

c. and shooting an image, and realizing accurate distance measurement between the train and the obstacle according to the relative position of the number on the image and the camera and the deflection angle.

In the invention, the distance measurement formula in step c is as follows:

wherein, alpha is the pitch angle, H is the camera height, theta is the vertical visual angle field, and the horizontal visual angle is beta, and Dmin is the image base and is apart from camera actual distance, and Dmax image topside is apart from camera actual distance, and height is the image height, and coordinate point X, Y are the horizontal distance and the vertical distance apart from the camera.

Example (b):

as shown in fig. 2, according to the above flow: reading in an image, graying the image, image adaptive median filtering, linear detection, distance prediction, detecting an actual track, reading in a video image from a camera, and detecting two tracks on the left side in the image as shown in a first image shown in fig. 2; firstly, preprocessing an image, graying the image, and obtaining a second picture according to an arrow; then, the grayed image containing noise is subjected to adaptive median filtering, as shown in the third diagram of fig. 2; further defining a region of interest in the image, as shown in the fourth diagram of fig. 2; performing pixel-based line detection on the region of interest, as shown in the fifth diagram of fig. 2; two straight lines are fitted and superimposed on the original as shown in the sixth diagram of fig. 2.

In conclusion, the method can be used for detecting the iron rail line, the algorithm comprises image preprocessing, line detection, line fitting and distance prediction, and experiments show that the line detection based on the visual pixel linking method has wide applicability.

It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.

Claims (6)

1. A train reversing auxiliary detection method based on visual pixel link line detection is characterized in that: the method comprises the following steps:

A. graying the original image;

B. carrying out denoising and filtering processing on the image of the gray level image, and processing the image polluted by noise by adopting a filtering method;

C. drawing an interest area in a picture to perform linear detection, so that a required linear line can be acquired more accurately;

D. detecting straight lines in an ROI area of the picture, filtering out horizontal straight lines and shorter straight lines, and taking only the longest line segment on each track;

E. extracting two detected paths and fitting the paths to an original image;

F. and detecting the distance between the obstacle and the lens.

2. A train reversing auxiliary detection method based on visual pixel link line detection is characterized in that: the image graying method in the step A is as follows:

a. respectively acquiring the energy of each color channel of the color image;

b. obtaining the optimal coefficient of each color channel according to the energy of each color channel;

c. and obtaining the gray level image corresponding to the color image according to the optimal coefficient of each color channel.

3. The train reversing auxiliary detection method based on the visual pixel link line detection according to claim 1, characterized in that: and in the step B, adaptive median filtering is adopted to filter out noise.

4. The train reversing auxiliary detection method based on the visual pixel link line detection according to claim 1, characterized in that: the specific method of the step F is as follows:

a. calibrating and correcting the distortion of the camera;

b. deducing the corresponding relation between an image coordinate system and a physical coordinate system according to the imaging principle of the optical lens of the camera;

c. and shooting an image, and realizing accurate distance measurement between the train and the obstacle according to the relative position of the number on the image and the camera and the deflection angle.

5. The train reversing auxiliary detection method based on the visual pixel link line detection according to claim 4, characterized in that: the distance measurement formula in the step c is as follows:

6. the train reversing auxiliary detection method based on the visual pixel link line detection according to claim 5, characterized in that:

wherein, alpha is the pitch angle, H is the camera height, theta is the vertical visual angle field, and the horizontal visual angle is beta, and Dmin is the image base and is apart from camera actual distance, and Dmax image topside is apart from camera actual distance, and height is the image height, and coordinate point X, Y are the horizontal distance and the vertical distance apart from the camera.

Images