A tire pattern analysis method and system based on image analysis

Abstract

The application discloses a kind of based on image analysis's tire pattern analysis method and system, it is related to tread pattern detection technical field.The present application discloses a kind of based on image analysis's tire pattern analysis method and system, it is related to tread pattern detection technical field.The present application discloses a kind of based on image analysis's tire pattern analysis method and system, it is related to tread pattern detection technical field.The present application discloses a kind of based on image analysis's tire pattern analysis method and system, it is related to tread pattern detection technical field.The present application discloses a kind of based on image analysis's tire pattern analysis method and system, it is related to tread pattern detection technical field.The present application discloses a kind of based on image analysis's tire pattern analysis method and system, it is related to tread pattern detection technical field.The present application discloses a kind of based on image analysis's tire pattern analysis method and system, it is related to tread pattern detection technical field.The present application discloses a kind of based on image analysis's tire pattern analysis method and system, it is related to tread pattern detection technical field.The present application discloses a kind of based on image analysis's tire pattern analysis method and system, it is related to tread pattern detection technical field.The present application discloses a kind of based on image analysis's tire pattern analysis method and system, it is related to tread pattern detection technical field.The present application discloses a kind of based on image analysis's tire pattern analysis method and system, it is related to tread pattern detection technical field.The present application discloses a kind of based on image analysis's tire pattern analysis method and system, it is related to tread pattern detection technical field.The present application discloses a kind of based on image analysis's tire pattern analysis method and system, it is related to tread pattern detection technical field.The present application discloses a kind of based on image analysis's tire pattern analysis method and system, it is related to tread pattern detection technical field.The present application discloses a kind of based on image analysis's tire pattern analysis method and system, it is related to tread pattern detection technical field.The present application discloses a kind of based on image analysis's tire pattern analysis method and system, it is related to tread pattern detection technical field.The present application discloses a kind of based on image analysis's tire pattern analysis method and system, it is related to tread pattern detection technical field.The present application discloses a kind of based on image analysis's tire pattern analysis method and system, it is related to tread pattern detection technical

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of tire pattern analysis, in particular to a tire pattern analysis method and system based on image analysis. BACKGROUND

[0002] Tire tread pattern is the pattern design of the part of the tire that contacts the road surface, and its main functions include drainage, skid resistance, enhanced grip, and noise reduction. Different types of patterns are suitable for different road conditions. The production process of tire tread pattern mainly includes design, mold manufacturing, and rubber molding. When tire tread pattern appears, it may affect the performance and safety of the tire, leading to decreased grip, which may cause skidding or loss of control, and affect drainage and grip, posing a safety hazard. Therefore, it is very important to detect tire tread pattern.

[0003] The patent with publication number CN115908304A discloses a system and method for detecting tire tread pattern of automobile tire. The system acquires 3D point cloud data of tire tread pattern in the detection area through an acquisition device. The host computer receives and processes the 3D point cloud data of tire tread pattern and determines whether the tire tread pattern exists. Thus, the system can efficiently detect the type and quantity of tire tread pattern of automobile tire, and accurately segment the shape of tire tread pattern, with low cost.

[0004] The above and similar technical solutions detect the tire tread pattern of the tire tread. When a single light source or a single angle is used for detection, the tire tread is a curved surface, and different tread patterns will result in different grooves on the tread. The curved surface object and the grooves will cause local overexposure or shadow in the image due to light reflection at different angles, which will cover up the small tread pattern or local tread pattern. This will further cause the fine cracks, contaminants, and irregular wear in the tire surface pattern to be difficult to detect. In addition, the tire tread pattern will not be the same in different motion states. However, when the tire is subjected to motion effect, dynamic blur will occur, causing the camera to capture a blurred image, which will affect the detection effect of the tread pattern. SUMMARY

[0005] The present application aims to provide a tire pattern analysis method and system based on image analysis to solve the problems raised in the background.

[0006] To achieve the above object, the present application provides the following technical solutions: a tire pattern analysis method based on image analysis, comprising: setting a multi-light source module, obtaining tire tread pattern feedback images of a target tire under the multi-light source module, obtaining a feedback image set, the multi-light source module including white light, ultraviolet light and infrared light, and the feedback image set including white light feedback images, ultraviolet light feedback images and infrared light feedback images;

[0007] Based on the white light feedback images, ultraviolet light feedback images and infrared light feedback images, first re-image preprocessing is performed respectively to obtain a first processed image set;

[0008] Based on the multi-light source module, a light source detection initial value and a light source detection limit value are set, a light source detection range is obtained based on the light source detection initial value and the light source detection limit value, the light source detection range is split to obtain a split range set;

[0009] Based on the split range set, the multi-light source module is adjusted to obtain a white light adjustment range, an ultraviolet light adjustment range and an infrared light adjustment range respectively, the light source intensity of the multi-light source module is changed respectively, the feedback image set updated image data is obtained after the light source intensity is changed, and the white light feedback image set, the ultraviolet light feedback image set and the infrared light feedback image set are obtained;

[0010] Based on the white light feedback image set, the ultraviolet light feedback image set and the infrared light feedback image set, second re-image preprocessing is performed respectively to obtain an updated feedback image set;

[0011] A judgment threshold is set, the judgment threshold is used to represent the recognition result range of the tire pattern and the recognition result comparison, the first processed image set and the updated feedback image set are classified and judged based on the judgment threshold, the classification feedback data of the first processed image set and the updated feedback image set is obtained, and then the actual judgment result of the tire tread pattern feedback images of the target tire under the multi-light source module is obtained.

[0012] Further, the method for obtaining the first processed image set comprises:

[0013] A reflectivity calibration component is set, light source brightness data of the white light feedback images, the ultraviolet light feedback images and the infrared light feedback images are obtained respectively to obtain white light brightness items, ultraviolet light brightness items and infrared light brightness items, and a brightness information set is obtained;

[0014] Based on the brightness information set, a brightness mean value is obtained, and the image brightness of the white light feedback images, the ultraviolet light feedback images and the infrared light feedback images is adjusted respectively based on the brightness mean value to obtain the first processed image set;

[0015] Further, the method for obtaining the light source detection range comprises:

[0016] Based on the white light module, a first intensity range is set, the first intensity range is an illumination intensity range, a standard illumination is detected, and a first intensity item is obtained, the standard illumination is an effective image acquisition range of the white light module when irradiating the tire to be detected;

[0017] Based on the ultraviolet light module, a second intensity information is set, the second intensity information is a waveband data information, a second intensity item is obtained, and the second intensity item is the best presentation of the stain image of the tire to be detected under the fixed waveband irradiation of the ultraviolet light module;

[0018] Based on the infrared light module, a third intensity range is set, the third intensity range is an irradiance intensity range, a third intensity item is obtained, and the third intensity item is the heat difference when the tire to be detected is irradiated by the infrared light module;

[0019] The first intensity item, the second intensity item, and the third intensity item are combined to obtain a light source detection range.

[0020] Further, the method for obtaining the split range set comprises:

[0021] A split item is set, the split item is an equal proportion item, the intensity difference between the starting intensity and the ending intensity of the first intensity item and the third intensity item is obtained, and a first difference item and a third difference item are obtained;

[0022] Based on the combination results of the split item and the first difference item and the third difference item, a first adjustment item and a third adjustment item are obtained;

[0023] Based on the combination of the first adjustment item and the third adjustment item with the starting intensity of the first intensity item and the third intensity item, a first combination set and a second combination set are obtained;

[0024] The first combination set, the second intensity item, and the third combination set are combined to obtain the split range set.

[0025] Further, the method for obtaining the white light feedback image set, the ultraviolet light feedback image set, and the infrared light feedback image set comprises:

[0026] Based on the white light adjustment range, the white light value in the multi-light source module is adjusted, the tire pattern feedback images under different white light values are obtained respectively, and a white light feedback image set is obtained;

[0027] Based on the ultraviolet light adjustment range, the ultraviolet light value in the multi-light source module is adjusted, the tire pattern feedback images under different ultraviolet light values are obtained respectively, and an ultraviolet light feedback image set is obtained;

[0028] Based on the infrared light adjustment range, the infrared light value in the multi-light source module is adjusted, the tire pattern feedback images under different infrared light values are obtained respectively, and an infrared light feedback image set is obtained.

[0029] Further, the setting method of the determination threshold comprises:

[0030] Obtaining the pattern depth information of the tire to be detected to obtain a target depth item, setting a first pair of mark values, and obtaining a first early warning value based on the combination result of the first pair of mark values and the target depth item;

[0031] Obtaining the outer surface information of the tire to be detected to obtain a target area item, setting a second pair of mark values, and obtaining a second early warning value based on the combination result of the second pair of mark values and the target area item;

[0032] Setting a temperature difference limit value to obtain a third early warning value, and combining the first early warning value, the second early warning value and the third early warning value to obtain the determination threshold.

[0033] Further, a tire pattern analysis system based on image analysis uses the tire pattern analysis method based on image analysis described above, comprising:

[0034] The acquisition module sets a multi-light source module, acquires the tire pattern feedback image of the target tire under the multi-light source module, and obtains a feedback image set, the multi-light source module includes white light, ultraviolet light and infrared light, and the feedback image set includes white light feedback image, ultraviolet light feedback image and infrared light feedback image;

[0035] The processing module performs first re-image preprocessing based on the white light feedback image, the ultraviolet light feedback image and the infrared light feedback image to obtain a first processing image set;

[0036] The splitting module sets a light source detection initial value and a light source detection limit value based on the multi-light source module, obtains a light source detection range based on the light source detection initial value and the light source detection limit value, splits the light source detection range to obtain a split range set, adjusts the multi-light source module based on the split range set to obtain white light adjustment range, ultraviolet light adjustment range and infrared light adjustment range respectively, changes the light source intensity of the multi-light source module respectively, acquires updated image data of the feedback image set after the light source intensity is changed to obtain white light feedback image set, ultraviolet light feedback image set and infrared light feedback image set;

[0037] The comparison module sets a determination threshold, the determination threshold is used to represent the recognition result range of the tire pattern and the recognition result comparison, classifies and judges the first processing image set and the updated feedback image set based on the determination threshold, acquires the classification feedback data of the first processing image set and the updated feedback image set, and further acquires the actual determination result of the tire pattern feedback image of the target tire under the multi-light source module.

[0038] Compared with the prior art, the beneficial effects of the present application are:

[0039] The tire pattern analysis method and system based on image analysis, by setting a multi-light source module, wherein the multi-light source module is a multi-light source camera combination module, including white light, ultraviolet light and infrared light, thereby realizing the irradiation effect of the multi-light source, obtaining the tire tread pattern detection data of the tire to be detected through the feedback result of the multi-light source, and performing image recognition analysis on the obtained tire to be detected in multiple range comparison, so that the tire tread pattern analysis effect is more comprehensive.

[0040] Meanwhile, in the case that the tire to be detected is fixed, the multi-light source module is driven to rotate around the tire to be detected, thereby realizing the effect of tire tread pattern detection of the tire to be detected at multiple angles on the basis of the multi-light source, and further improving the accuracy of the tire tread pattern analysis.

[0041] Furthermore, by driving the tire to be detected to rotate, the tire to be detected is in a motion state, at this time, the tire tread pattern of the tire to be detected in the motion state can be detected, multiple mode detection of the tire to be detected is realized, and the multi-light source module rotating simultaneously with the first module reduces the dynamic module of the tire to be detected as much as possible due to rotation, and improves the accuracy of the tire tread pattern analysis. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 It is a whole flowchart of the present application;

[0043] Figure 2 It is a composition diagram of the multi-light source module of the present application;

[0044] Figure 3 It is a split range flowchart of the present application;

[0045] Figure 4 It is a transverse telescopic module diagram of the present application;

[0046] Figure 5 It is a light module and tire C distance diagram of the present application;

[0047] Figure 6 It is a combination diagram of the reflective feedback image, ultraviolet light feedback image and infrared light feedback image of the present application. DETAILED DESCRIPTION

[0048] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0049] Precise tire tread detection not only ensures the safety performance of the tire, but also prolongs the service life of the tire, thereby reducing resource consumption and environmental pollution. However, due to the complexity of the tire tread itself, and the influence of light and motion state in the detection process, the detection method of single light source or single angle has significant limitations in the tire tread pattern detection of the tire tread. The tire tread is not a plane, but a curved surface with a certain curvature. This curved surface characteristic, combined with the diversity of the tire tread pattern, causes various grooves on the tire tread. When light shines on these curved surfaces and grooves, complex reflection phenomena occur. Reflection of light at different angles can cause overexposure in local areas of the image, i.e. high brightness, loss of details; while in other areas, shadows can occur, resulting in low brightness, which also hides the details. This uneven lighting greatly interferes with the identification of small tire tread patterns, such as micro-cracks in the tire tread pattern, attached contaminants, and irregular wear, which are often hidden by the influence of light and difficult to detect. In addition, the presentation of the tire tread pattern will change under different motion states of the tire. For example, micro-cracks that are difficult to observe under static conditions may open up due to stress when the tire is rolling, making them more visible. Therefore, in order to more comprehensively detect the tire tread pattern, it is necessary to detect the tire tread pattern under the motion state of the tire. However, applying motion to the tire can cause dynamic blur when the camera captures images due to the high-speed rotation of the tire. This blurring effect makes the image details unclear, further reducing the accuracy of the tire tread pattern detection. The tire pattern analysis system based on image analysis provided by the present application sets a multi-light source module, which is a multi-light source camera combination module, including white light, ultraviolet light, and infrared light, thereby achieving the illumination effect of multiple light sources. The tire tread pattern detection data of the tire to be detected is obtained through the feedback results of the multiple light sources, making the tire tread pattern detection effect more comprehensive. At the same time, under the condition that the tire to be detected is fixed, the multi-light source module is driven to rotate around the tire to be detected, thereby achieving the effect of detecting the tire tread pattern of the tire to be detected at multiple angles based on multiple light sources, further improving the accuracy of the tire tread pattern detection. Finally, the tire to be detected is driven to rotate, so that the tire to be detected is in a motion state. At this time, the tire tread pattern of the tire to be detected in the motion state can be detected, realizing multi-mode detection of the tire to be detected. The multi-light source module rotating simultaneously with the tire to be detected minimizes the dynamic module of the tire to be detected due to rotation, improving the accuracy of the tire pattern analysis, as shown in Figure 1

[0050] Step S100: Set a multi-light source module to obtain a tire tread pattern feedback image of a target tire under the multi-light source module.

[0051] It should be noted that, as​Figure 6 As shown, the target tire is obtained under the feedback image set of the multi-light source module, the multi-light source module includes white light, ultraviolet light and infrared light, and the feedback image set includes white light feedback image, ultraviolet light feedback image and infrared light feedback image.

[0052] Step S200: Based on the white light feedback image, the ultraviolet light feedback image and the infrared light feedback image, the first image preprocessing is carried out respectively.

[0053] It should be noted that after the first image preprocessing, the first processed image set is obtained, and the acquisition method of the first processed image set includes: setting a reflectivity calibration component, acquiring light source brightness data of the white light feedback image, the ultraviolet light feedback image and the infrared light feedback image respectively, obtaining white light brightness item, ultraviolet light brightness item and infrared light brightness item, and obtaining brightness information set; Based on the brightness information set, the brightness average value is obtained, and the image brightness of the white light feedback image, the ultraviolet light feedback image and the infrared light feedback image is adjusted based on the brightness average value, and the first processed image set is obtained.

[0054] Step S300: Set the light source detection initial value and the light source detection limit value, and obtain the light source detection range.

[0055] It should be noted that, as shown in Figure 2 and Figure 6 As shown, the acquisition method of the light source detection range includes: based on the white light module, setting the first intensity range, the first intensity range is the illumination intensity range, acquiring the detection standard illumination, obtaining the first intensity item, and the detection standard illumination is the scratch effective image acquisition range when the white light module irradiates the tire to be detected; Based on the ultraviolet light module, set the second intensity information, the second intensity information is the wave band data information, obtain the second intensity item, the second intensity item is the best presentation of the stain image of the tire to be detected under the fixed wave band irradiation of the ultraviolet light module; Based on the infrared light module, set the third intensity range, the third intensity range is the irradiance intensity range, obtain the third intensity item, the third intensity item is the heat difference when the infrared light module irradiates the tire to be detected; The first intensity item, the second intensity item and the third intensity item are combined to obtain the light source detection range.

[0056] In the specific implementation process, it is now necessary to detect the pattern on the surface of a certain tire by light irradiation, and white light, ultraviolet light and infrared light are selected for irradiation. First of all, according to the industrial detection standard illumination, the intensity range of white light is 10000lux-30000wlux. When the illumination is too low, i.e.​When the intensity is 30000 lux, the curved reflection is overexposed, which can cover the depth information, and the first intensity range is 10000 lux-30000 lux. Secondly, the second intensity information is the fixed waveband information of ultraviolet light. When the waveband is too low, the fluorescent signal of chemical pollutants such as silane residue is weak and not easy to observe. When the waveband information is high, it is easy to accelerate the aging of rubber and produce background fluorescence interference. Therefore, the selected waveband information is 365nm waveband. Finally, the third intensity range is 500W / m 2 -1500W / m 2 When the intensity is <500W / m 2 , the internal delamination temperature difference anomaly is difficult to detect, and when the intensity is >1500W / m 2 , the surface rubber is easy to overheat and deform, producing thermal artifacts. Therefore, the selected irradiance intensity range is 500W / m 2 -1500W / m 2 The final light source detection range is: the intensity range of white light is 10000 lux-30000 lux, the waveband information of ultraviolet light is 365nm waveband, and the irradiance intensity range of infrared light is 500W / m 2 -1500W / m 2 .

[0057] Step S400: split the light source detection range to obtain a split range set.

[0058] It should be noted that, as Figure 3 indicated, the method for obtaining the split range set comprises: setting a split item, the split item is an equal proportion item, which is 50%, obtaining the intensity difference between the start intensity and the end intensity of the first intensity item and the third intensity item to obtain the first difference item and the third difference item; based on the combination results of the split item and the first difference item and the third difference item, the first adjustment item and the third adjustment item are obtained; based on the combination of the start intensity of the first intensity item and the third intensity item with the first adjustment item and the third adjustment item, the first combination set and the second combination set are obtained; the first combination set, the second intensity item and the third combination set are combined to obtain the split range set.

[0059] In the specific implementation process, the surface defects of a certain tire need to be detected by light irradiation. White light, ultraviolet light and infrared light are selected for irradiation. The selected light source detection range is: the intensity range of white light is 10000 lux-30000 lux, the waveband information of ultraviolet light is 365nm waveband, and the irradiance intensity range of infrared light is 500W / m 2 -1500W / m 2 , i.e. the first intensity item 10000 lux-30000 lux, the third intensity item 500W / m 2 1500W / m 2, the first difference term is 20000 lux, and the third difference term is 1000 W / m 2 , according to the combination result of the set split term 50% and the first difference term and the third difference term, the first adjustment term is 10000 lux, and the third adjustment term is 500 W / m 2 , at this time, the combination result of the first adjustment term and the third adjustment term with the starting intensity of the first intensity term and the third intensity term is 20000 lux and 1000 W / m respectively 2 , the first combination set is 10000 lux, 20000 lux, 30000 Wlux, and the third combination set is 500 W / m 2 , 1000 W / m 2 , 1500 W / m 2 , at this time, the split range set is 10000 lux, 20000 lux, 30000 Wlux, 365nm waveband, 500 W / m 2 , 1000 W / m 2 , 1500 W / m 2 .

[0060] Step S500: Adjusting the multi-light source module based on the split range set, changing the light source intensity of the multi-light source module, and obtaining updated image data of the feedback image set after the light source intensity is changed.

[0061] It should be noted that the updated image data of the feedback image set after the light source intensity is changed obtains the white light feedback image set, the ultraviolet light feedback image set and the infrared light feedback image set, and the method for obtaining the white light feedback image set, the ultraviolet light feedback image set and the infrared light feedback image set includes: adjusting the white light value in the multi-light source module based on the white light adjustment range, obtaining the tire pattern feedback image under different white light values respectively, and obtaining the white light feedback image set; adjusting the ultraviolet light value in the multi-light source module based on the ultraviolet light adjustment range, obtaining the tire pattern feedback image under different ultraviolet light values respectively, and obtaining the ultraviolet light feedback image set; adjusting the infrared light value in the multi-light source module based on the infrared light adjustment range, obtaining the tire pattern feedback image under different infrared light values respectively, and obtaining the infrared light feedback image set.

[0062] Step S600: Set a determination threshold to classify and determine the first processing image set and the updated feedback image set.

[0063] It should be noted that the determination threshold is used to represent the identification result range of the tire pattern and the identification result comparison, the classification feedback data of the first processing image set and the updated feedback image set are obtained, and then the actual determination result of the tire pattern feedback image of the target tire under the multi-light source module is obtained. The setting method of the determination threshold includes: obtaining the pattern depth information of the tire to be detected, obtaining the target depth item, setting the first pair of mark value, the first pair of mark value is 30%, based on the combination result of the first pair of mark value and the target depth item, obtaining the first warning value; obtaining the surface information of the tire to be detected, obtaining the target area item, setting the second pair of mark value, the second pair of mark value is 25%, based on the combination result of the second pair of mark value and the target area item, obtaining the second warning value; setting the temperature difference limit value, the temperature difference limit value is ±2℃, obtaining the third warning value, the first warning value, the second warning value and the third warning value are combined to obtain the determination threshold.

[0064] In the specific implementation process, it is necessary to obtain the image irradiation of a certain tire b by a module a composed of white light, ultraviolet light and infrared light, the tire pattern depth of the tire b is 10mm, the surface area of the tire b is 0.4m 2 , at this time, according to the set first pair of mark value 30%, the first warning value is 3mm, according to the second pair of mark value, the second warning value is 0.1m 2 , according to the temperature difference limit value ±2℃, the third warning value is obtained, the tire is irradiated by white light, ultraviolet light and infrared light respectively, and the related information of the tire is obtained, it is obtained that the tire b has 10 scratches, the depths are 0.5mm, 0.3mm, 0.2mm, 0.2mm, 0.3mm, 0.3mm, 0.3mm, 0.2mm, 0.2mm, 0.3mm respectively, at this time, the cumulative scratch depth is 2.8mm, which does not reach the set first warning value 3mm; it is obtained that the tire has 3 stains, the areas are 0.03m 2 , 0.02m 2 , 0.02m 2 , the cumulative stain area is 0.07m 2 , which does not reach the set second warning value 0.1m 2 ; the tire is uniformly irradiated by infrared light, the feedback temperature information is obtained, the set infrared light is 1000W / m 2 , the temperature rise information of the irradiation position is 5.4℃, 5.5℃, 5.5℃, 5.8℃, 6.2℃, 5.6℃, 5.7℃, 6.7℃, 5.7℃, 5.2℃, which does not reach the set third warning value ±2℃.

[0065] It should be noted that, by setting the first module for placing the tire to be detected, and setting the multi-light source module to revolve around the first module, the first module can revolve, setting the first detection speed and the second detection speed, setting the first initial detection speed and the second initial detection speed, obtaining the first initial speed term and the second initial speed term; based on the first initial speed term, the tire to be detected is rotated, and it is judged whether the tire to be detected shakes; when the tire to be detected shakes, a decreasing ratio is set, the decreasing ratio is 10%, the first initial speed term is decreased based on the decreasing ratio, until the tire to be detected does not shake, the first detection speed is obtained; based on the second initial speed term, the tire to be detected is rotated, and it is judged whether the tire to be detected shakes; when the tire to be detected does not shake, an increasing ratio is set, the increasing ratio is 10%, the second initial speed term is increased based on the increasing ratio, until the tire to be detected shakes, the second detection speed is obtained, and the position of the multi-light source module is adjusted, based on the first detection speed, the first module and the tire to be detected are rotated.

[0066] It should be noted that, as shown in Figure 4 , the multi-light source module includes a transverse telescopic module and a light module, the image acquisition range of the multi-light source module is obtained, the effective range term is obtained, the multi-light source module is adjusted based on the transverse telescopic module, the position of the light module when the interactive range of the effective range term and the tire to be detected is maximum is obtained, the marker position is obtained, and then the adjustment of the position of the multi-light source module is realized.

[0067] In the specific implementation process, as shown in Figure 5 , it is now necessary to irradiate and acquire images of a certain tire by a module A composed of white light, ultraviolet light and infrared light, place the tire C on the rotatable module B, rotate the module A around the module B, obtain the effective range of the white light in the light module of the module A as 80°, the effective range of the ultraviolet light as 80°, and the effective range of the infrared light as 90°, at this time the comprehensive effective range of the light module in the module A is 80°, the effective range term is obtained, at this time the transverse adjustment module A adjusts the distance between the light module and the tire C, when the effective range term of the light module and the interactive range of the tire C are at the maximum, the position of the light module in the module A is determined, at this time the distance between the light module and the tire C is 0.288m.

[0068] It should be noted that, at the first detection speed, the first module and the tire to be detected are rotated, the camera module acquires the marker position of the tire to be detected, which is set as the marker initial point, the length information between the marker initial points is acquired, the acquisition range term is obtained, the acquisition range term is also the distance between the marker points, that is, the distance when the tire to be detected rotates one circle, that is, the circumference of the tire to be detected; based on the split range set, the white light module, the ultraviolet light module and the infrared light module acquire the scratch information, the fluorescent information of the pollutants and the infrared temperature difference information of the tire to be detected based on the acquisition range term.

[0069] It should be noted that the first module and the multi-light source module are rotated at the second detection speed at the same time, the surface image data of the tire to be detected is obtained through the detection method, after the image information of the surface image data of the tire to be detected is obtained, based on the set determination threshold, when there is more than the determination threshold in the image information, the tire pattern detection feedback data is output, the detection method comprises: obtaining the image acquisition range of the multi-light source module, based on the interaction information of the acquisition range and the tire to be detected, obtaining the mapping range item, the mapping range item is used to represent the range of the image acquisition range of the multi-light source module falling on the tire to be detected; based on the mapping range item, the tire to be detected is divided to obtain at least two division result items, when the mapping range item is half of the area of the tire, the division result item is two at this time, the start position information and the end position information of the division result item are obtained, at least two division start items and division end items are obtained; the multi-light source module obtains the surface image data of the tire to be detected in different split range sets in the first division result item, to obtain a second image information subset; the speed of the multi-light source module is lowered, so that the division start item of the division result item in the order coincides with the division end item of the first division result item, the surface image data of the tire to be detected in different split range sets in the division result item in the order is repeatedly obtained through the multi-light source module, to obtain the image information.

[0070] Although the embodiments of the present application have been shown and described, it can be understood by those skilled in the art that various changes, modifications, replacements and variations can be made to the embodiments without departing from the principles and spirits of the present application, the scope of the present application is defined by the appended embodiments and their equivalents.

Claims

1. A tire pattern analysis method based on image analysis, comprising: setting a multi-light source module, obtaining a tire tread pattern feedback image of a target tire under the multi-light source module, obtaining a feedback image set, the multi-light source module including white light, ultraviolet light and infrared light, the feedback image set including white light feedback images, ultraviolet light feedback images and infrared light feedback images; characterized in that: based on the white light feedback images, the ultraviolet light feedback images and the infrared light feedback images, first re-image preprocessing is performed respectively to obtain a first processed image set; based on the multi-light source module, setting a light source detection initial value and a light source detection limit value, obtaining a light source detection range based on the light source detection initial value and the light source detection limit value, splitting the light source detection range to obtain a split range set; based on the split range set, adjusting the multi-light source module to obtain white light adjustment range, ultraviolet light adjustment range and infrared light adjustment range respectively, changing the light source intensity of the multi-light source module respectively, obtaining updated image data of the feedback image set after the light source intensity is changed to obtain white light feedback image set, ultraviolet light feedback image set and infrared light feedback image set, and then obtaining an updated feedback image set; setting a judgment threshold, the judgment threshold being used to represent the recognition result range of the tire pattern and the recognition result comparison, classifying and judging the first processed image set and the updated feedback image set based on the judgment threshold, obtaining classification feedback data of the first processed image set and the updated feedback image set, and then obtaining the actual judgment result of the tire tread pattern feedback image of the target tire under the multi-light source module; the method for obtaining the light source detection range comprises: based on the white light module, setting a first intensity range, the first intensity range being an illumination intensity range, obtaining a detection standard illumination to obtain a first intensity term, the detection standard illumination being a scratch effective image acquisition range when the white light module irradiates the tire to be detected; based on the ultraviolet light module, setting a second intensity information, the second intensity information being waveband data information, obtaining a second intensity term, the second intensity term being a best presentation of a stain image of the tire to be detected under fixed waveband irradiation of the ultraviolet light module; based on the infrared light module, setting a third intensity range, the third intensity range being an irradiance intensity range, obtaining a third intensity term, the third intensity term being a heat difference when the infrared light module irradiates the tire to be detected; the first intensity term, the second intensity term and the third intensity term are combined to obtain the light source detection range; the method for obtaining the split range set comprises: setting a split term, the split term being an equal proportion term, obtaining intensity difference values of start and end intensities of the first intensity term and the third intensity term to obtain first and third difference terms; based on the combination results of the split term and the first and third difference terms respectively, obtaining first and third adjustment terms; based on the combination of the first and third adjustment terms and the start intensities of the first and third intensity terms respectively, obtaining first and second combination sets; the first combination set, the second intensity term and the third combination set are combined to obtain the split range set.

2. A method of analyzing tire pattern based on image analysis according to claim 1, characterized in that: the method for obtaining the first processed image set comprises: The reflectivity calibration component is configured to obtain light source brightness data of the white light feedback image, the ultraviolet light feedback image, and the infrared light feedback image respectively, to obtain a white light brightness term, an ultraviolet light brightness term, and an infrared light brightness term, and to obtain a brightness information set; Based on the brightness information set, a brightness mean value is obtained, and based on the brightness mean value, the image brightness of the white light feedback image, the ultraviolet light feedback image, and the infrared light feedback image is adjusted respectively to obtain a first processed image set.

3. A method of analyzing tire pattern based on image analysis according to claim 1, characterized in that: The setting method of the determination threshold value includes: Obtain the pattern depth information of the tire to be detected to obtain a target depth term, set a first pair of mark values, and based on the combination result of the first pair of mark values and the target depth term, obtain a first warning value; Obtain the outer surface information of the tire to be detected to obtain a target area term, set a second pair of mark values, and based on the combination result of the second pair of mark values and the target area term, obtain a second warning value; Set a temperature difference limit value to obtain a third warning value, and the first warning value, the second warning value, and the third warning value are combined to obtain the determination threshold value.

4. An image analysis based tire pattern analysis system characterized by: A tire pattern analysis method based on image analysis is used, including: The acquisition module is configured to set a multi-light source module, acquire a tire tread pattern feedback image of a target tire under the multi-light source module, and obtain a feedback image set, the multi-light source module including white light, ultraviolet light, and infrared light, and the feedback image set including a white light feedback image, an ultraviolet light feedback image, and an infrared light feedback image; The processing module is configured to perform first re-image preprocessing based on the white light feedback image, the ultraviolet light feedback image, and the infrared light feedback image to obtain a first processed image set; The splitting module is configured to set a light source detection initial value and a light source detection limit value based on the multi-light source module, obtain a light source detection range based on the light source detection initial value and the light source detection limit value, split the light source detection range to obtain a split range set, adjust the multi-light source module based on the split range set, obtain a white light adjustment range, an ultraviolet light adjustment range, and an infrared light adjustment range respectively, change the light source intensity of the multi-light source module respectively, obtain updated image data of the feedback image set after the light source intensity is changed, and obtain a white light feedback image set, an ultraviolet light feedback image set, and an infrared light feedback image set; The comparison module is configured to set a determination threshold value, the determination threshold value is used to represent the recognition result range of the tire pattern and the recognition result comparison, classify and judge the first processed image set and the updated feedback image set based on the determination threshold value, obtain classification feedback data of the first processed image set and the updated feedback image set, and further obtain an actual determination result of the tire tread pattern feedback image of the target tire under the multi-light source module.

Images