A Method for Collecting Indicator Light Status Information of a Traffic Control Equipment Cabinet

Abstract

The present invention relates to a method for collecting status information of indicator lights in a traffic control equipment cabinet, including the following steps: collecting and preprocessing an image of the traffic control equipment cabinet; performing a semi-Hough transform on the image to calibrate the coordinates of the indicator lights in the equipment cabinet; obtaining the actual color gamut of each indicator light display, capturing the moment when the indicator light changes, and obtaining the display color of the indicator light.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent operation and maintenance of rail transit, and relates to a method for collecting the status information of the indicator lights of traffic control equipment cabinets. Background Art

[0002] The operation and maintenance inspection of rail transit is a typical labor-intensive industry, mainly relying on manual inspection, handwritten recording of various information, formulating maintenance plans based on experience, and regularly carrying out inspections. Its disadvantages are high inspection intensity, low efficiency, high labor cost, and being affected by the skylight point, making it difficult to meet the sustainable development needs of large-scale urban rail transit networks. However, at present, due to the rapid development of urban rail transit, the number of trains has increased sharply, and the operating conditions have become increasingly complex. As a result, the train operation interval has been shortened, and the end-of-operation time has been extended, which has brought great pressure to maintenance and guarantee.

[0003] At present, the indicator lights of the equipment cabinets in the existing machine rooms of rail transit operation and maintenance facilities still rely on manual inspection and manual recording and reporting. Usually, there are hundreds of indicator lights in a machine room. The manual inspection operation method is backward, the degree of automation is low, the workload is huge, and there is a lack of data accumulation and intercommunication, which is far from the vision of achieving full coverage of inspection, intelligent automatic identification of faults, and providing feasible solutions in the development of intelligent operation and maintenance. Summary of the Invention

[0004] Based on this, in view of the existing situation, the present invention proposes a method for collecting the status information of the indicator lights of traffic control equipment cabinets, replacing manual inspection and enabling relevant personnel to grasp the status information of the indicator lights of traffic control equipment cabinets in the rail transit field in real time.

[0005] To achieve the above object, the solution of the present invention is: collecting the image of the equipment cabinet and preprocessing it; performing semi-Hough transform on the image to calibrate the coordinates of the indicator lights of the equipment cabinet; obtaining the actual color gamut in different display states of the indicator lights, capturing the moment when the indicator light changes, and obtaining the display color of the indicator light.

[0006] Further, in the step (S2), performing semi-Hough transform on the image to calibrate the coordinates of the indicator lights of the equipment cabinet specifically includes the following steps:

[0007] (S21) Manually roughly judge the position area of the indicator lights in the image. In order to reduce the computational complexity of the semi-Hough transform, first roughly delimit the position of the indicator lights, that is, the fuzzy position parameters of the indicator lights are (x′ i , y′ i , r′ i ), where the serial number i ∈ [0, N), N is the number of indicator lights, x′ i , y′ i, r′ represent the abscissa of the fuzzy center, the ordinate of the fuzzy center, and the fuzzy radius respectively. Let the accurate center coordinates and radius of the indicator light be (x i , y i , r i ). The selection of the fuzzy position parameters of the indicator light should satisfy:

[0008]

[0009] And set the maximum radius R, satisfying R ≥ r′ i , r i .

[0010] (S22) Use the position parameters (x′ i , y′ i , r′ i ) in (S21) to perform a semi-Hough transform on the preprocessed edge map to obtain the accurate position parameters (x i , y i , r i ). Crop the edge map to obtain the rectangular region of interest according to the position parameters (x′ i , y′ i , r′ i ). The center coordinates of the rectangle are (x′ i , y′ i ), and the side lengths are all 2r′ i to obtain the i-th edge sub-image I i (m, n), where the row number m ∈ [0, K i , the column number n ∈ [0, K i , and K i represents the pixel side length of the edge sub-image.

[0011] Find the position parameters x i and y i . When m = m0, if there exists n = n0 that satisfies the non-edge point (m0, n0), and there is an edge point (m0, n -1 ) and (m0, n1) on each side of its left and right, and n0 - n -1 = n1 - n0, then perform a semi-Hough transform on the point (m0, n0), that is:

[0012] m0cosθ = ρ

[0013] Obtain the Hough coordinates (θ, ρ), traverse θ ∈ [-π / 2, π / 2] to obtain the parameter space of ρ when m = m0. Traverse m to obtain the parameter space of ρ, perform a frequency count on ρ, and the m corresponding to the ρ with the largest frequency is the position parameter x i . Similarly, traverse n to obtain the position parameter y i .

[0014] Find the position parameter r i , substitute the edge point (m, n) into the equation to obtain the candidate radius r″ i；(m,n) , perform a semi-Hough transform, that is:

[0015] r″cosθ = ρ

[0016] Obtain the Hough coordinates (θ, ρ), traverse θ ∈ [-π / 2, π / 2] to obtain the parameter space of ρ, perform a frequency count on ρ, and the r″ corresponding to the ρ with the largest frequency i；(m,n) , is the position parameter r″ i .

[0017] Furthermore, in the step (S3), take the actual color gamut in different display states of the indicator light, capture the moment when the indicator light changes, and obtain the display color of the indicator light, which specifically includes the following steps:

[0018] (S31) According to the pixel radius and center coordinates of the indicator light, obtain the actual color gamut in different display states of the indicator light. For the convenience of calculation, define the display area of the i-th indicator light as a rectangle centered at (x i , y i ) and with as the side length, denoted by S(m, n, c). Among them, m and n are the row and column numbers respectively, and c is the channel number. The indicator light can display H colors in total. Manually make the indicator light light up the h-th color (not limited to three colors of red, green, and blue), and calculate the actual color gamut vector V(h), that is:

[0019]

[0020] (S32) Extract the display area G(m, n) of the i-th indicator light from the preprocessed grayscale image, take the difference from the same display area G′(m, n) in the previous frame, and calculate the change amount ΔG, that is:

[0021]

[0022] When a given threshold ΔG0 is set, if ΔG > ΔG0, it is considered that the display of the indicator light has changed, and continue to judge the display color of the indicator light; if ΔG ≤ ΔG0, it is considered that the display of the indicator light has not changed, wait and re-perform this step.

[0023] (S33) When the display of the i-th indicator light changes, obtain its color gamut V i , that is:

[0024]

[0025] Solve for the serial number h such that (V(h) - V i ) 2The smallest, and the color corresponding to the serial number h is the color of the display of the i-th indicator light. Brief Description of the Drawings

[0026] Figure 1 is the image of the indicator light of the equipment cabinet involved in the present invention;

[0027] Figure 2 is the block diagram of the working principle of the present invention. Detailed Embodiment

[0028] The following combines the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0029] For the convenience of understanding by those skilled in the art, an embodiment is now used to collect and preprocess the image of the equipment cabinet: the camera is fixedly installed facing the indicator light of the equipment cabinet, the image resolution is 1920*1080, and the preprocessing performed includes but is not limited to grayscale conversion, Gaussian equalization, and edge detection. The original image, grayscale image, and edge image are respectively stored in the variable matrices I, G, and E for convenient use in subsequent steps.

[0030] For the convenience of understanding by those skilled in the art, an embodiment is now used to perform semi-Hough transform on the image to calibrate the coordinates of the indicator light of the equipment cabinet and make an explanation:

[0031] Specifically, in the step (S21), there are 326 indicator lights in the picture. Taking the 5th indicator light as an example, the fuzzy position parameters are (40, 43, 15).

[0032] Specifically, in the step (S22), using the position parameters (40, 43, 13) in (1), perform semi-Hough transform on the edge image E. The obtained rectangular region of interest has the center coordinates of the rectangle as (40, 43) and the side length of 26 pixel units. The accurate position parameters obtained through semi-Hough transform are (42, 45, 10).

[0033] For the convenience of understanding by those skilled in the art, an embodiment is now used to obtain the display color gamut of the indicator light in different display states, capture the moment when the indicator light changes, and obtain the display color of the indicator light to make an explanation:

[0034] Specifically, in the step (S41), the center of the display rectangle area of the 5th indicator light is (42, 45), and the side length is 7. The indicator light can display 3 colors, namely red, green, and blue, corresponding to h equal to 1, 2, and 3 respectively. The calculated real color gamut vectors are: V(1) = (243 139 125); V(2) = (103 245 114); V(3) = (98 107 246).

[0035] Specifically, in the step (S42), when the threshold value ΔG0 of the differential processing is 2.45.

[0036] Specifically, in the step (S43), when the display of the 5th indicator light changes, its color gamut V5 = (240 134 129) is calculated. When h is 1 (V(h) - V i ) 2 is the smallest, so the display color of the 5th indicator light is red.

Claims

1. A method for collecting the status information of the indicator lights of a traffic control equipment cabinet, characterized in that, Including the following steps: S1. Collect the image of the equipment cabinet and preprocess it; S2. Perform semi-Hough transform on the image to calibrate the coordinates of the equipment cabinet indicator lights, including the following steps: S21. Manually determine the position area of the indicator light in the image; set the fuzzy position parameters of the indicator light as (x i ′, y i ′, r i ′), where the serial number i ∈ [0, N), N is the number of indicator lights, x i ′, y i ′, r′ respectively represent the abscissa of the fuzzy center, the ordinate of the fuzzy center, and the fuzzy radius; set the accurate center coordinates and radius of the indicator light as (x i , y i , r i ). The selection of the fuzzy position parameters of the indicator light should satisfy: And set a maximum radius R, satisfying R≥r i ′,r i ; S22. Use the indicator position area in S21 to perform a semi-Hough transform on the edge map to obtain accurate position parameters (x i , y i , r i ); According to the position parameters (x i ′, y i ′, r i ′), crop the edge map to obtain the rectangular region of interest. The center coordinates of the rectangle are (x i ′, y i ′), and the side lengths are both 2r i ′, to obtain the i-th edge sub-map I i (m, n), where the row number m ∈ [0, K i , the column number n ∈ [0, K i , and K i represents the side length of the edge sub-image pixels; when m = m0, if there exists n = n0 such that the non-edge point (m0, n0) has an edge point (m0, n -1 ) and (m0, n1) on each of its left and right sides, and n0 - n -1 = n1 - n0, then perform a semi-Hough transform on the point (m0, n0) as m0cosθ = ρ to obtain the Hough coordinates (θ, ρ). Traverse θ ∈ [-π / 2, π / 2] to obtain the parameter space of ρ when m = m0; traverse m to obtain the parameter space of ρ, perform a frequency count on ρ, and the m corresponding to the ρ with the largest frequency is the position parameter x i ; Similarly, traverse n to obtain the position parameter y i ; Substitute the edge point (m, n) into the equation to obtain the candidate radius r″ i；(m,n) , perform a semi-Hough transform r″cosθ = ρ to obtain the Hough coordinates (θ, ρ). Traverse θ ∈ [-π / 2, π / 2] to obtain the parameter space of ρ, perform a frequency count on ρ, and the r″ i；(m,n) corresponding to the ρ with the largest frequency is the position parameter r i ″; S3. Obtain the actual color gamut of the indicator lights in different display states, capture the moment when the indicator lights change, and obtain the display color of the indicator lights, including the following steps: S31. According to the pixel radius and center coordinates of the indicator lights in S2, obtain the actual color gamut of the indicator lights in different display states; S32. Determine whether the display of the indicator lights has changed between the current frame and the previous frame; if it has changed, proceed to S33 to continue determining the color of the indicator lights; if there is no change, return to S32 and wait for the display of the indicator lights to change; S33. Obtain the color gamut of the indicator lights in the current frame, compare it with the actual color gamut in S31, and obtain the display color of the current indicator lights.

Images