CN107369177B - A warning method for preventing over-tensioning of the winch rope of road rescue equipment based on graphic recognition - Google Patents

Abstract

The invention discloses a warning method for over-pull prevention of a winch rope of road rescue equipment based on graphic recognition, comprising the following steps: 1. Uniformly marking a geometric figure on the circumference of the outer wheel disc of the winch rope groove; 2. A graphic acquisition device is arranged on the outside of the groove, and the acquisition area of the graphic acquisition device is larger than a single geometric figure mark, and the area larger than a single geometric figure mark is set in the collection area as an area of interest, so that the interest area can only contain one geometric figure mark at most ; 3. Identify the type of geometric figure markers included in the region of interest in each frame of image collected by the graphics acquisition device; 4. Continuously identify the collected images to obtain a sequence of geometric figure marker changes; change the geometric figure markers Sequence to obtain the rotation direction and rotation angle of the winch shaft; 5. Calculate the rotation direction and rotation angle of the winch shaft, and give a safety alarm when the number of winch coils is less than the threshold.

Description

A warning method for preventing over-tensioning of the winch rope of road rescue equipment based on graphic recognition

technical field

The invention belongs to the field of safety early warning of road rescue equipment, and in particular relates to an early warning method for preventing over-drawing of a winch rope.

Background technique

With the rapid growth of my country's economy, road traffic infrastructure conditions are gradually improving, and the number of motor vehicles is also increasing year by year. Social demands such as road vehicle accident rescue, urban illegal vehicle handling, and natural disaster rescue and rescue are also rapidly increasing. At the same time, the roadside assistance industry, which is the aftermarket of automobiles, is gradually being valued and has developed rapidly. According to relevant data, during the operation of road rescue equipment, there are often situations where road rescue equipment cannot perform effective rescue. One of them is that the attached winch cannot achieve effective pulling and lifting operations. The reason is that, on the one hand, there are factors such as insufficient performance of the road rescue equipment itself, such as insufficient traction capacity of the winch; on the other hand, the relevant monitoring equipment for road rescue equipment has not been perfected, and cannot provide operators with accurate information about related equipment, such as the winch The uncertainty of the degree of rope retraction is also an unavoidable factor; but at present, there is no research on the related technology of monitoring the degree of retraction and retraction of the winch rope attached to the winch of the road rescue equipment.

The winch is an important part of road rescue equipment, and it is a traction device, which is mainly used for pulling and hoisting the rescued vehicle; during road rescue operations, the winch rope wound on the winch drum should not be less than 10 turns to ensure the friction between the winch rope and the drum to provide tension for the winch rope, otherwise there will be a dangerous situation of insufficient pull force of the winch rope or even the rope head falling off and the winch rope breaking. The control of the retraction and retraction of the winch rope in the traditional winch mainly relies on manual observation. This method is easily affected by factors such as bad weather and different viewing angles, and has many shortcomings.

Contents of the invention

Purpose of the invention: Aiming at the problems existing in the prior art, the present invention discloses a warning method for over-tensioning the winch rope of road rescue equipment based on graphic recognition. Regular geometric marks are set on the outside and identified by template matching algorithm to realize real-time monitoring of the winch rope retraction degree and avoid the situation of over-tensioning the winch rope, so as to achieve the purpose of improving the rescue efficiency and safety performance of road rescue equipment.

Technical solution: The present invention adopts the following technical solution: a graphic recognition-based early warning method for preventing over-tension of the winch rope of road rescue equipment, including the following steps:

(1) Uniformly mark geometric figures on the circumference of the outer wheel of the winch rope groove;

(2) Graphic acquisition device is set outside the road rescue equipment winch rope groove, the acquisition area of said graphic acquisition device is greater than a single geometric figure mark, and the area greater than a single geometric figure mark is set in the acquisition area as the area of interest, so that the interested A region can contain at most one geometry marker;

(3) identifying the type of geometric figure markers contained in the region of interest in each frame of image collected by the graphics acquisition device;

(4) Continuously identify the images collected to obtain the geometric figure mark change sequence; according to the geometric figure mark change sequence, obtain the rotation direction and the rotation angle of the winch rotating shaft;

(5) Calculate the rotation direction and rotation angle of the winch shaft, and give a safety alarm when the number of winch coils is less than the threshold.

The graphic acquisition device can adopt a low-cost camera with an output frequency of 25 Hz, and its acquisition frequency supports a winch with a maximum rope speed of 10 m/min.

A lighting device is provided on the graphics collection device, and the lighting device is used for illuminating the collection area of the graphics collection device to deal with road rescue scenes with insufficient light.

There are three kinds of geometric figures, numbered 0, 1, 2 respectively; one group is in the order of 012, and 4 groups are evenly marked on the circumference of the outer wheel disc of the winch rope groove.

When the geometric figure mark change sequence is 021, 102, 210, the winch shaft rotates clockwise; when the geometric figure mark change sequence is 012, 120, 201, the winch shaft rotates counterclockwise.

The three geometric figures are square, circle and triangle respectively.

In step 3, the template matching method is used to identify the geometric mark type in the image region of interest.

Beneficial effects: Compared with the prior art, the graphic recognition-based winch rope over-tension warning method for road rescue equipment disclosed by the present invention has the following advantages: 1. The equipment required for the implementation of the method is simple and low in cost, and template matching is used for identification Geometric graphics, simple algorithm, fast recognition speed, meeting the maximum rope speed of the winch rope; 2. Combined with auxiliary lighting equipment, it can meet the application in road rescue scenes with insufficient light.

Description of drawings

Fig. 1 is the schematic diagram of the geometric figure marking outside the capstan rope groove baffle plate in the embodiment;

Fig. 2 is a schematic diagram of the installation position of the pattern recognition device;

Fig. 3 is a schematic diagram of template matching method for recognizing geometric figure markers.

Detailed ways

The present invention will be further explained below in conjunction with the accompanying drawings and specific embodiments.

A graphic recognition-based early warning method for preventing over-tension of a winch rope for road rescue equipment, comprising the following steps:

(1) Uniformly mark geometric figures on the circumference of the outer wheel of the winch rope groove;

There are three kinds of geometric figure marks in this embodiment: square "□", circle "○" and equilateral triangle "△"; the three are evenly arranged on the roulette in the clockwise order of "□○△". It is set in the form of four groups a week, that is, each geometric figure mark occupies the angle range of 30° of the wheel, and a total of four groups of 12 geometric figure marks are carried out in one week. The specific marking form is shown in Figure 1. Three geometric figures The color of the mark adopts the contrasting color of the background color of the wheel to improve the accuracy of identification. In this embodiment, the color of the winch rope groove wheel is red, so the color of the geometric figure mark is selected as green.

(2) Graphic acquisition device is set outside the road rescue equipment winch rope groove, the acquisition area of said graphic acquisition device is greater than a single geometric figure mark, and the area greater than a single geometric figure mark is set in the acquisition area as the area of interest, so that the interested A region can contain at most one geometry marker;

As shown in Figure 2, a graphic acquisition device 4 is arranged at a fixed position outside the winch rope groove of the road rescue equipment. This embodiment selects a low-cost camera with an output frequency of 25 Hz, and the acquisition frequency supports a winch whose maximum rope speed is 10 m/min. In addition, lighting auxiliary devices are installed to deal with road rescue scenes with insufficient light.

The acquisition area 1 of the graphics acquisition device is the position of the roulette facing the camera. Adjust the camera position or parameters to make the acquisition area 1 larger than the size of a single geometric mark 3, and regard the rectangular area larger than a single geometric mark as interested Area 2, the specific form is shown in Figure 1 and Figure 2. In Fig. 2, the winch rope is wound on the swivel bearing 6, and the swivel bearing 6 and the winch rope baffle plate 5 rotate together when the rope is released or taken in, and the geometric figure marks collected by the camera are also changing.

(3) identifying the type of geometric figure markers contained in the region of interest in each frame of image collected by the graphics acquisition device;

The area of interest of the graphic acquisition device is slightly larger than a single geometric figure mark, and can only contain one geometric figure mark at most, that is, for the region of interest of each frame of image, the pattern recognition device can perform at most one type of geometric figure mark For identification, this embodiment uses a similarity measure method to perform template matching on the region of interest in each frame of image collected by the image acquisition device.

The geometric pattern templates in this embodiment are square "□", circle "○" and equilateral triangle "△" respectively, and the template matching algorithm is used to judge whether the image data of the region of interest collected by the graphic acquisition device contains marked The geometric pattern template and the mathematical model of the matching algorithm are shown in Figure 3. The operation idea is as follows: the existing geometric pattern templates, namely the square "□", the circle "○" and the equilateral triangle "△" are sequentially obtained from the collected Starting from the starting position of the pixel of the image data, compare it with an area of the same size, and then shift to the next pixel, and then continue to perform the same operation until all positions in the image are compared, and the similarity measure reaches the preset threshold. Indicates that the image data collected by the graphics acquisition device contains the corresponding geometric figure template, otherwise, the recognition result is regarded as "empty".

According to the document "Application of Template Matching in Image Recognition", formula (1) can be used to measure the similarity between the collected image data and the geometric template:

In the formula (1), T represents the geometric figure template, and k represents the numbering of the geometric figure template, which is coded as [0, 1, 2] in the present embodiment to represent the square template "□", the circular template "○" and Equilateral triangle template "△"; S represents the image data collected in the area covered by the geometric figure template, that is, the sub-image, and i, j represent the coordinates of the pixel in the upper left corner of the sub-image S in the collected image data, that is, the reference point, where The value range is: 1<i,j<N-M+1, N is the side length pixel of the region of interest, M is the side length pixel of the geometric template; D(i,j) indicates that the reference point is (i ,j) the similarity degree of the subgraph with respect to the corresponding geometric figure template.

Expand the formula (1) to get:

In formula (2), the first term after the equal sign represents the energy of the corresponding subgraph S _{i, j} covered by the geometric template, which changes with the position of the reference point (i, j); the second term represents the template and The relationship between the subgraphs also changes with the position of the reference point (i, j), and the value of this item reaches the maximum when the two are completely matched; the third item is the total energy of the geometric template, which is a The value has nothing to do with the position of the reference point (i, j); therefore, formula (3) can be used as a similarity measure:

In formula (3), R(i,j) is the measure of the similarity between the geometric pattern template and the collected image data. When the two are completely matched, R(i,j)=1, otherwise R(i,j) <1; Obviously, the larger R(i,j), the more similar the geometric pattern template is to the collected image data.

(4) Continuously identify the images collected to obtain the geometric figure mark change sequence; according to the geometric figure mark change sequence, obtain the rotation direction and the rotation angle of the winch rotating shaft;

The recognition results of each frame of image data are square template "□", circular template "○", equilateral triangle template "△" and no matching template "empty"; since the acquisition frequency of the image recognition device is 25 Hz, that is, every One frame of image data is collected in 40ms, and the rotation speed of the road rescue equipment with the winch shaft is relatively slow, so there will be a situation where the same geometric figure mark is obtained by collecting several consecutive frames. The change sequence of the geometric figure mark in this embodiment will be the same continuously The result output is regarded as a single recognition result output, and for the situation that the recognition result is "empty", the change sequence of the geometric figure mark in this embodiment is not included in the processing; every time the geometric figure in the recognition result changes, it indicates that the winch shaft Rotated by 30°.

The possible situation of the change sequence of geometric figure marks is as follows: (1) Queue 1: □△○, coded as [0,2,1]; (2) Queue 2: ○□△, coded as [1,0,2] ;(3) Queue 3: △○□, coded as [2,1,0]; (4) Queue 4: □○△, coded as [0,1,2]; (5) Queue 5: ○△□ , coded as [1,2,0]; (6) Queue 6: △□○, coded as [2,0,1]; where “□”, “○” and “△” represent the recognized square template, Circular template and equilateral triangle template.

The aforementioned 6 situations are all possible sequences of changes in geometrical figures in this embodiment, and each sequence appears once to represent that the winch shaft has rotated 0.25 circles; combined with Figure 1, it can be seen that sequences 1, 2, and 3 are clockwise of the winch shaft Rotating state, that is, the winch is in the state of releasing the rope; sequence 4, 5, 6 is the state of the winch shaft rotating counterclockwise, that is, the winch is in the state of receiving the rope.

(5) Calculate the rotation direction and rotation angle of the winch shaft, and give a safety alarm when the number of winch coils is less than the threshold.

Knowing the initial value of the number of winch coils, through the judgment and calculation of the rotation direction and rotation angle of the winch shaft in steps 1-4, the remaining coils of the winch rope can be calculated and monitored in real time, and when the calculated number of winch coils is less than the threshold Security alarm. In this embodiment, the threshold is set to 10 circles.

After the above steps, the system proposed by the present invention can effectively monitor the degree of retraction and retraction of the winch rope of the road rescue equipment, so as to achieve the purpose of improving the rescue efficiency and safety performance of the road rescue equipment.

Claims (7)

1. A road rescue equipment winch rope anti-over-pulling early warning method based on graphic recognition, is characterized in that, comprises the steps: (1) Uniformly mark geometric figures on the circumference of the outer wheel of the winch rope groove; (2) Graphic acquisition device is set outside the road rescue equipment winch rope groove, the acquisition area of said graphic acquisition device is greater than a single geometric figure mark, and the area greater than a single geometric figure mark is set in the acquisition area as the area of interest, so that the interested A region can contain at most one geometry marker; (3) identifying the type of geometric figure markers contained in the region of interest in each frame of image collected by the graphics acquisition device; (4) Continuously identify the images collected to obtain the geometric figure mark change sequence; according to the geometric figure mark change sequence, obtain the rotation direction and the rotation angle of the winch rotating shaft; (5) Calculate the rotation direction and rotation angle of the winch shaft, and give a safety alarm when the number of winch coils is less than the threshold. 2. The road rescue equipment winch rope anti-overtension early warning method based on graphic recognition according to claim 1, wherein the graphic acquisition device is a camera with an output frequency of 25 Hz. 3. the road rescue equipment winch rope anti-over-pulling early warning method based on graphic recognition according to claim 1, is characterized in that, lighting equipment is set on the graphics collection device, and the lighting equipment is used for the collection area of the graphics collection device illumination. 4. The road rescue equipment winch rope anti-overdrawing early warning method based on graphic recognition according to claim 1, characterized in that, there are three kinds of geometric figures, respectively numbered 0, 1, 2; the order of 012 is one Mark 4 groups evenly on the circumference of the outer wheel of the winch rope groove. 5. The road rescue equipment winch rope anti-overdrawing early warning method based on graphic recognition according to claim 1, characterized in that, in step 3, a template matching method is used to identify the geometric figure mark type in the region of interest of the image. 6. The road rescue equipment winch rope anti-over-pulling early warning method based on graphic recognition according to claim 4, characterized in that, when the change sequence of the geometric figure mark is 021, 102, 210, the winch shaft rotates clockwise; When the change sequence of the geometric figure mark is 012, 120, 201, the winch shaft rotates counterclockwise. 7 . The graphic recognition-based early warning method for preventing over-tensioning of the winch rope of road rescue equipment according to claim 4 , wherein the three geometric figures are square, circle and triangle respectively. 8 .