CN113781545B - A method for rapid identification of geometrical features of irregular particles - Google Patents

Abstract

The invention discloses a method for rapidly identifying geometrical features of irregular particles. Cloud image; the fragments and the standard size are taken together by the camera, and after converting the point cloud image, the pixel grid of the point cloud image occupied by the standard size is observed, and the actual size ratio of each pixel grid is obtained after comparison; the three-axis length characterization algorithm is used to obtain fragments The three semi-major axes under the three-view, and the three semi-major axes are used as the basic data to define the fragmentation degree FR, elongation coefficient EC, flatness coefficient FC, and sphericity S of the fragments. The invention provides a method for quickly identifying the geometric features of irregular particles, which can quickly scan the fragments and obtain basic data such as relatively accurate size characterization quantities, and use a length characterization algorithm to obtain the triaxial length values of the fragments, which can effectively process The time can be shortened to 1/3 of the existing technology.

Description

A method for rapid identification of geometrical features of irregular particles

technical field

The invention relates to a representational identification method of graphics. More particularly, the present invention relates to a method for rapidly identifying geometrical features of irregular particles for use in the analysis of post-impact fragments.

Background technique

The phenomenon of fragmentation generated by crushing widely exists in nature and human production and life. The fragments produced by materials under the action of impact, blasting, weathering, grinding and compression are of different sizes and shapes. Understanding the crushing process and the size and shape of the resulting fragments is of great significance to security defense, blasting mining, gradation design, and milling efficiency. The shape of debris has attracted much attention in space exploration, landform evolution, mining and excavation, rockfill dam filling, impact crushing, and powder grinding. Understanding the shape of debris is beneficial to assess the impact of space debris on spacecraft and satellites, reveal the weathering and spalling, wear and collapse processes of rock debris, and control the energy demand and debris quantity for powder grinding. For rock engineering, it can improve rock crushing efficiency and reduce energy waste. , to produce fragments with controllable particle size and shape, and save the cost of secondary processing.

Under dynamic loads, a large number of fragments will be generated during the material crushing process. Limited by the test method and measurement technology, it takes a lot of time to scan and identify each fragment. Few studies have paid attention to the shape of the fragments after particle crushing. The comprehensive characterization of fragments is still Difficulties and challenges.

3D scanning imaging technology has been used to study the deformation and fracture of various materials. Through 3D digital image data, quantitative research on particles and fragments formed by crushing is carried out. By using 3D scanning to obtain the initial shape of the fragments, using 3D scanning to scan the geometric shape of the fragments, applying digital image processing technology to characterize the shape of the fragments, and exploring the shape characteristics of the fragments after a single object is broken, it is helpful to analyze the initial shape of the particles. The impact of the particle size distribution and shape characteristics is studied to provide a certain reference for the crushing process and debris phenomenon from the perspective of particle and debris shape, and to better understand the crushing mechanism and crushing form under dynamic loads.

SUMMARY OF THE INVENTION

SUMMARY OF THE INVENTION An object of the present invention is to address at least the above-mentioned problems and/or disadvantages and to provide at least the advantages that will be described hereinafter.

To achieve these objects and other advantages in accordance with the present invention, a method for rapidly identifying geometrical features of irregular particles is provided, comprising:

In step 1, particle size classification is performed on the crushed particles through a classification system;

In step 2, the fragments at all levels are respectively placed in the scanning system for three-dimensional scanning, so as to obtain three views corresponding to each fragment and size information related to the fragments;

Step 3, by performing image data processing on the three views to obtain the corresponding point cloud map;

Step 4: By comparing the actual standard size in the point cloud image with the pixel points occupied by the standard size in the point cloud image, the actual size ratio of each pixel point is obtained;

Step 5: Based on the actual size ratio of each pixel, the three-axis length representation algorithm is used to obtain the three semi-major axes of the fragments in the three views, and the three semi-major axes are used as the basic data to define the fragmentation degree FR, extension of the fragments. Length factor EC, flatness factor FC, sphericity S.

Preferably, the grading system is configured to include:

control module;

Vibration grading assembly arranged at the channel opening under the anti-sputtering collection device;

A multi-stage conveyor belt arranged under the vibration transmission assembly;

Sorting platform matched with conveyor belts at all levels;

At least one manipulator arranged on one side of the sorting platform to send the graded fragments to the scanning system for three-dimensional scanning;

wherein the vibration grading assembly is configured to include:

A vibrating plate with a slope greater than 10 degrees and a power mechanism matched with it are provided with screening openings with multi-stage apertures, and a protective plate with an arc-shaped structure is provided on the edge of the vibrating plate;

a first sensor arranged on the vibrating plate;

The conveyor belts at all levels are provided with matching second sensors;

A matching third sensor is arranged on the sorting platform.

Preferably, in step 1, the method of particle size classification treatment is configured to include:

S10, trigger the first sensor when the debris falls on the vibration plate, the first sensor transmits the acquired first signal to the control module, and the control module switches the working state of the power mechanism based on the received first signal, so that the vibration plate is in the working state ;

S11, the debris falling into the vibrating plate is classified according to the external size under the continuous vibration of the vibrating plate and the action of the gradient and the screening port, and the classified debris is sent to the corresponding conveyor belt;

S12, the second sensor is triggered by the debris falling on the conveyor belt, the second sensor transmits the acquired second signal to the control module, and the control module switches the working state of the conveyor belt based on the received second signal, so that the debris of the conveyor belt is transported to the control module. The sorting platform realizes the graded processing of debris.

Preferably, in step 2, the third sensor is triggered by the debris falling into the sorting platform, the third sensor transmits the acquired third signal to the control module, and the control module switches the working state of the manipulator based on the received third signal, In order to send the fragments of the sorting platform to the scanning system respectively for 3D scanning operation;

The scanning system is configured to include:

Scanning platform with matching black background plates in three dimensions;

A micro-focus camera that cooperates with the background plate to take three-view images of the debris in three dimensions;

A processing module that communicates with each camera.

Preferably, the processing module first preprocesses the image, and then uses an image enhancement algorithm to determine the clarity of the currently captured picture;

The processing module selects a first position in the image that can be used to adjust the focus based on the judgment result by judging the image sharpness for the first time;

The processing module compares and analyzes the image sharpness again to select a second position in the image that can be used to adjust the focus based on the analysis result, and repeats this step to obtain the corresponding lens focus position.

Preferably, in step five, the length characterization algorithm is configured to include:

Based on the three views, construct three ellipsoids whose semi-major axes are a, b, and c in turn, and specify that a≥b≥c. Assuming that in one of the views, the pixels occupied by the fragments are n, and the side length of each pixel is λ, based on the area equivalence principle, each semi-major axis can be obtained based on the following formulas:

Preferably, it also includes performing splicing and reorganization processing on the three-dimensionally scanned pictures, and the splicing and reorganization processing method is configured to include:

Based on the surface profile of the fragments obtained after scanning, and the surface profile information of the fragments recorded by the scanning system, the cross-sections of each fragment are compared and combined through data processing to determine whether they can be matched, so that the object before the impact is restored.

Preferably, the method for comparing the sections and judging the combination is configured to include:

Taking the plane where the two parallel highest and lowest points of the fragment cross section are located as the base plane, and taking the lowest and highest points of the cross section as the base points for calculating the volume, the main volume and missing volume of the fracture surface are calculated by calculating;

Among them, the main volume is the real volume calculated from the lowest point to the highest point of the section, and the missing volume is the virtual volume calculated from the highest point to the lowest point of the section. By comparing and analyzing whether the volumes overlap, the matching fragment surfaces can be connected and reconstructed. If there are unmatched fragments, they can be alienated and then matched again.

The present invention includes at least the following beneficial effects: firstly, the present invention is mainly to quickly scan the fragments and obtain basic data such as relatively accurate size representations; The hardware involved in scanning, scanning, etc. can be controlled by the computer, and then the data measured by the hardware can be processed by algorithms to obtain basic parameters such as size representation, and the resulting parameters can be displayed by the computer.

Second, the present invention mainly focuses on the scanning processing of debris and the integration of fast channels. In the process flow, particle size screening, debris transmission, and debris scanning are all automatic and integrated operations. When the debris reaches the corresponding processing position, the trigger is triggered. The sensor enables the device to operate and use its functions without changing the location, and it can operate quickly and step by step, which can effectively improve the processing speed. In terms of data, the length characterization algorithm is mainly used for processing. The required three-view images are obtained by scanning the geometric features of the fragments. The scanning process does not require manual focusing processing, and then image data processing is performed on the scanned three-view images. The point cloud image occupied by the fragment image is obtained, and then the three-axis length value of the fragment is quickly and directly obtained through the algorithm, which is used for other experimental studies.

Third, the processing of fragments in the prior art requires that the three views taken are converted into point cloud images, and then the point cloud images are calculated. It takes a lot of time to manually record the number of pieces, so the processing time of one piece usually takes at least 10 minutes, and the present invention calculates the relevant size representation parameters of pieces through algorithm calculation, and it takes about 2-3 minutes, so a piece of piece processing It can save 7-8 minutes, so the processing time of one fragment data can be shortened to 1/3 of the existing technology. At the same time, because there is not only one fragment when processing fragments, but many fragments need to be processed, it can be seen that this The technology as a whole can save a lot of time and improve efficiency to a large extent.

Other advantages, objects, and features of the present invention will appear in part from the description that follows, and in part will be appreciated by those skilled in the art from the study and practice of the invention.

Description of drawings

1 is a process flow diagram of a method for rapidly identifying geometric features of irregular particles in an embodiment of the present invention;

Fig. 2 is the original picture taken by the camera in the scanning system;

Fig. 3 is the point cloud image obtained after analyzing Fig. 2 by adopting the processing method of the present invention to rapidly identify the geometric feature of irregular particles;

FIG. 4 is a schematic diagram of the structural layout of the grading system of the present invention.

Detailed ways

The present invention will be further described in detail below with reference to the accompanying drawings, so that those skilled in the art can implement it with reference to the description.

An implementation form of a method for rapidly identifying geometric features of irregular particles according to the present invention, including:

In step 1, the crushed particles are subjected to particle size classification processing by the classification system. In the actual processing, for the measurement of the fragments, the fragments can also be measured with a vernier caliper with an accuracy of 0.02mm as the standard actual measurement value. , and then use the micro-focus camera to measure the debris as the measurement value. The minimum value of the micro-focus can reach 0.2mm. The actual value measured by the vernier caliper and the measurement value measured by the micro-focus camera are compared and analyzed as the debris-related size information. basic data to control the error rate of measurement results to less than 20%;

In step 2, the fragments at all levels are placed in the scanning system for 3D scanning, so as to obtain three views corresponding to each fragment and size information related to the fragments. The characterization quantities and surface parameters of the material are used as basic data for other studies. For example, when calculating the fragmentation degree under dynamic loading, relevant data information such as fragment size is required, and basic data can also be provided for the study of the fragmentation process and failure mechanism. The invented hierarchical scanning does not include the scanning of small fragments at all levels, and the small fragments of each level are directly regarded as spheres of the same diameter and will not be scanned;

Step 3, by performing image data processing on the three views to obtain the corresponding point cloud map, in practical applications, the point cloud data can be generated after importing the image into the program;

Step 4: By comparing the actual standard size in the point cloud image with the pixel points occupied by the standard size in the point cloud image, the actual size ratio of each pixel point is obtained. After entering the program and converting it into point cloud data, it has the corresponding fragment point cloud map and ruler point cloud map. The ruler in the point cloud map has the corresponding point cloud pixel grid. By comparing the actual scale scale with the point cloud pixel grid, you can Get the actual size of the pixel grid, and then analyze the parameters of the fragment;

Step 5: Based on the actual size ratio of each pixel, the three-axis length representation algorithm is used to obtain the three semi-major axes of the fragments in the three views, and the three semi-major axes are used as the basic data to define the fragmentation degree FR, extension of the fragments. Length factor EC, flatness factor FC, sphericity S;

In actual operation, based on the situation that a large amount of debris will be generated after the impact, the workload of debris processing is relatively large, and a large amount of debris needs to be classified and measured, and larger debris needs to be scanned. Compared with manual processing, the process of processing fragments is faster, more efficient and smarter, thereby reducing time costs. It can quickly, accurately and automatically process fragments, obtain relevant experimental data such as size characterization, surface parameters, etc., and provide experimental data for subsequent experimental research on fragmentation and failure characteristics, which can better study the failure process and failure mechanism. Specifically, , In practical applications, the existing technology takes too long to process fragments because there is no streamlined processing method, and it takes about 30 minutes to process all fragments by using the algorithm of this technology to process fragments. The reason is that the present invention, through the combined processing of hardware and software, is based on the automatic integration process, so that the automatic fragment classification and scanning of geometric features can be carried out very quickly, and the elapsed time in each link is shortened, so that The processing time of the whole process is greatly reduced, and the efficiency is greatly improved.

As shown in Figure 1, the grading system of the present invention is configured to include:

control module (not shown);

The vibration grading assembly 2 is arranged at the channel opening under the anti-sputtering collection device 1;

A multi-stage conveyor belt 3 arranged under the vibration transmission assembly;

The sorting platform 4 matched with the conveyor belts at all levels;

at least one manipulator (not shown) arranged on one side of the sorting platform to send the graded fragments to the scanning system for three-dimensional scanning;

wherein the vibration grading assembly is configured to include:

The vibrating plate 5 with a slope greater than 10 degrees and the power mechanism (not shown) matched with it are provided with a screening port 6 with multi-stage apertures, and the edge of the vibrating plate is provided with a protection with an arc-shaped structure in section. board 7;

a first sensor (not shown) arranged on the vibrating plate;

Each level of conveyor belt is provided with a second sensor (not shown) which is matched;

A matching third sensor (not shown) is arranged on the sorting platform;

In actual operation, the method of particle size classification treatment is configured to include:

S10, trigger the first sensor when the debris falls on the vibration plate, the first sensor transmits the acquired first signal to the control module, and the control module switches the working state of the power mechanism based on the received first signal, so that the vibration plate is in the working state ;

S11, the debris falling into the vibrating plate is classified according to the external size under the continuous vibration of the vibrating plate and the action of the gradient and the screening port, and the classified debris is sent to the corresponding conveyor belt;

S12, the second sensor is triggered by the debris falling on the conveyor belt, the second sensor transmits the acquired second signal to the control module, and the control module switches the working state of the conveyor belt based on the received second signal, so that the debris of the conveyor belt is transported to the control module. The sorting platform realizes the classification of debris; in this scheme, the debris after impact falls into the vibrating screening device through the channel below the anti-sputtering collection device, and triggers the first sensor when it falls to the vibrating screening device below. , the signal is transmitted to the computer (control module), and the computer makes an instruction to open the vibrating screen;

The crushed particles are vibrated along the slope (slope of about 10°) and fall into the screening ports of different aperture sizes (0.2, 0.4, 0.8). The size of the fragments can be classified in the process. There should be a protective layer to prevent the fragments from splash when vibrating;

After being classified, fragments of different sizes fall to the conveyor belt below and trigger the second sensor to transmit a signal to the computer, and the computer instructs the motor control module to start the conveyor belt;

There is a pick-up area (sorting platform) at the end of the conveyor belt, and there is a robotic arm claw that picks up debris. When the debris reaches the pickup area, the sensor will be triggered, the signal will be transmitted to the computer, and the computer will give instructions to stop the conveyor belt and pick up the robotic arm claw. When the debris reaches the scanning area, the robot arm will automatically return to its position after picking it up. When there is no debris in the picking area, the conveyor belt will run automatically.

In another example, in step 2, the third sensor is triggered by the debris falling on the sorting platform, and the third sensor transmits the acquired third signal to the control module, and the control module switches the manipulator based on the received third signal. In the working state, the fragments of the sorting platform are respectively sent to the scanning system for three-dimensional scanning operation. The debris is scanned three-dimensionally. After the scanning, the computer instructs the rotation sensor to rotate the scanning platform, and then scans the debris particles in the other direction. After scanning, the computer instructs the computer to make instructions, and starts the robotic arm claw to pick up the debris on the platform to the collection device. Then pick up the fragments in the pick-up area on the conveyor belt to the scanning platform, and repeat this operation to identify the geometric characteristics of the remaining fragments, and obtain the outline of the fragments after processing and analysis by the 3D scanning system;

The scanning system is configured to include:

Scanning platform with matching black background plates in three dimensions;

A micro-focus camera that cooperates with the background plate to take three-view images of the debris in three dimensions. Multiple micro-focus cameras that cooperate with the background plate can be arranged on the scanning platform at the same time as needed, so that the debris can be captured without turning. The corresponding three views are obtained through the micro-focus camera;

a processing module connected in communication with each camera, the processing module first preprocesses the image, and then uses an image enhancement algorithm to determine the clarity of the currently captured picture;

The processing module selects a first position in the image that can be used to adjust the focus based on the judgment result by judging the image sharpness for the first time;

The processing module compares and analyzes the image clarity again to select a second position in the image that can be used to adjust the focus based on the analysis results, repeat this step to obtain the corresponding lens focus position, and classify the fragments according to the traditional method. As well as the characterization of the geometrical features of the fragments, the required basic parameters such as the mass, volume, and scale characterization of the fragments are obtained; the processing process of this traditional method requires the fragments to be sieved with a grade sieve to sieve the particle size of the fragments, and then compare the size of the fragments. Large particle size fragments are scanned and analyzed, which needs to be self-focused and then photographed, and then the basic data is obtained by hand calculation and processing. This process takes a lot of time;

However, the present invention adds an automatic focusing function to the scanning system; firstly, the image is preprocessed to compensate for uneven illumination, and to enhance the image detail information, and the image enhancement algorithm is used to obtain the clarity of the image from the image data. Adjust the focal length at the position with better clarity, then compare and analyze the image clarity, and then find the position with better clarity to focus, repeat this step, the image will change after each focus adjustment, so as to find the image At the clearest moment, the lens position corresponding to the clearest picture is the focus position. It provides high-quality image data for focus determination based on image clarity. When the debris is scanned, it can automatically focus according to the situation of the debris. Compared with manual manual focusing, the scanning process is faster and more accurate.

In another example, in step five, the length characterization algorithm is configured to include:

Based on the three views, construct three ellipsoids whose semi-major axes are a, b, and c in turn, and specify that a≥b≥c. Assuming that in one of the views, the pixels occupied by the fragments are n, and the side length of each pixel is λ, based on the area equivalence principle, each semi-major axis can be obtained based on the following formulas:

After each semi-major axis is obtained, the fracture degree FR, elongation coefficient EC, and flattening coefficient FC can be expressed by and formulas as needed:

Among them, V ₀ is the initial volume of the specimen,

The sphericity S can be expressed by the following formula:

In another example, it also includes performing splicing and recombination processing on the three-dimensionally scanned pictures, and the splicing and recombination processing method is configured to include:

Based on the surface profile of the fragments obtained after scanning, and the surface profile information of the fragments recorded by the scanning system, the cross-sections of each fragment are compared and combined through data processing to determine whether they can be matched, so that the object before the impact is restored.

In another example, the method of comparing and combining the sections is configured to include:

Taking the plane where the two parallel highest and lowest points of the fragment cross section are located as the base plane, and taking the lowest and highest points of the cross section as the base points for calculating the volume, the main volume and missing volume of the fracture surface are calculated by calculating;

Among them, the main volume is the real volume calculated from the lowest point to the highest point of the section, and the missing volume is the virtual volume calculated from the highest point to the lowest point of the section. By comparing and analyzing whether the volumes overlap, the matched fragment surfaces can be connected and reconstructed. If there are unmatched fragments, they can be alienated and then matched again.

Example:

After screening, the small-sized fragments are analyzed, and the smaller-sized fragments can be directly regarded as spheres with the same sieve size hole size, and the scanning processing of the smaller fragments can be ignored directly, and then the remaining larger Fragments are scanned and analyzed to calculate relevant characterization parameters;

As shown in Figure 1-3, taking the processing of one of the fragments as an example, the collected fragments are sieved and transferred to the platform, and the three perspectives of the fragments are photographed. During the shooting process, the fragments and the Shooting at a standard size in a nearby position, you can get the image of the standard size and the fragment in each view, with a black background as the base, you can get the camera prototype image of the fragment;

The three views of the fragment are processed by PS to obtain a more optimized image, which makes the black and white contrast of the image obvious;

The processed image is calculated with the three-axis length characterization algorithm program, and the image is imported into the calculation program, and the point cloud image with the same shape as the original image is obtained after conversion;

The fragment and the standard size placed in the nearby position are photographed by the micro-focus camera, and the image of the standard size and fragment in each view can be obtained. Compare, get the actual size ratio of each pixel point;

The method of equivalent circle area is used to characterize the triaxial length of the fragment characteristic ellipsoid, and three views are used to construct three ellipsoids whose semi-major axes are a, b, and c in turn, and stipulate that a≥b≥c, with a As an example to introduce the calculation method, it is assumed that in this view, the pixel grid occupied by the fragments is n, the side length of each pixel point is λ, and the area is equivalent, the calculation process of a is as follows

By calculating b and c in the same way, the three-axis representation length of the fragment can be obtained, and then the same processing method is performed on the three views of the remaining larger fragments, and the obtained length representation parameters such as a, b, and c are used as basic data. For other studies, geometric parameters such as fragmentation rate FR (Fragmentation Rate), elongation coefficient EC (Elongation Coefficient), flatness coefficient FC (Flatness Coefficient), and sphericity S (Sphericity) can be defined. The influence of each parameter on the morphology of the fragments is as follows: the larger the fragmentation FR, the higher the degree of fragmentation of the specimen; the smaller the EC, the closer the shape of the fragments is to the needle shape; the smaller the FC, the flatter the shape of the fragments, the closer to the cake shape ; the larger S is, the closer the shape of the fragment is to a sphere.

In the present invention, the particle size classification of the fragments is carried out by a screening device, and the fragments with different particle sizes are separated for mass, volume and other measurements to obtain the basic parameters of the fragments; The object, and then the remaining particle size fragments can be processed for the next step of analysis. Compared with the traditional method, it can save a lot of time, and the process has been improved, so that the user can use it very conveniently; when scanning, it can automatically focus and then scan the fragments. This focusing process is convenient for non-professionals to use, It can also greatly save time and make the picture more accurate; the three views obtained by scanning, after image processing, introduce a three-axis length characterization algorithm, and directly obtain the required three-axis length through the algorithm, just import the image. The parameters can be calculated, which improves the efficiency. The calculation and processing of the point cloud image makes the number of pixels accurate through the algorithm, and the whole process does not require manual calculation, which reduces the chance of errors and greatly reduces the time, so that the efficiency is improved. Through this integration Compared with the original method, the system has been greatly improved in processing methods, and the automated processing makes it more efficient and accurate.

The above solution is only an illustration of a preferred example, but not limited thereto. When implementing the present invention, appropriate substitutions and/or modifications may be made according to user needs.

The number of apparatuses and processing scales described here are intended to simplify the description of the present invention. Applications, modifications and variations to the present invention will be apparent to those skilled in the art.

Although embodiments of the present invention have been disclosed above, they are not limited to the applications set forth in the specification and embodiments. It can be fully adapted to various fields suitable for the present invention. Additional modifications can readily be implemented by those skilled in the art. Therefore, the invention is not to be limited to the specific details and illustrations herein shown and described, without departing from the general concept defined by the appended claims and the scope of equivalents.

Claims (4)

1. A method for rapidly identifying irregular particle geometries, comprising:

step one, carrying out particle size grading treatment on crushed particles through a grading system;

respectively placing the fragments at all levels in a scanning system to perform three-dimensional scanning so as to obtain three views corresponding to the fragments and size information related to the fragments, wherein scanning of small fragments at all levels is not included when the scanning system is used for performing graded scanning;

thirdly, processing image data of the three views to obtain a corresponding point cloud picture;

step four, comparing the actual standard size in the point cloud picture with the pixel points occupied by the standard size in the point cloud picture to obtain the actual size occupation ratio of each pixel point;

step five, based on the actual size ratio of each pixel point, three semimajor axes of the fragments under three views are obtained by adopting a triaxial length characterization algorithm, and the fragmentation degree FR, the elongation coefficient EC, the flatness coefficient FC and the sphericity S of the fragments are defined by taking the three semimajor axes as basic data;

in the fourth step, the scale and the fragments are shot on the same picture, and after the shot pictures are converted into point cloud data in a program, a corresponding fragment three-view point cloud picture and a scale point cloud picture are obtained, and the actual size of a pixel grid is obtained by comparing the actual scale with the pixel grid of the point cloud;

in the second step, a third sensor is triggered through the fragments falling into the sorting platform, the third sensor transmits an obtained third signal to the control module, and the control module switches the working state of the mechanical arm based on the received third signal so as to respectively send the fragments of the sorting platform into the scanning system for three-dimensional scanning operation;

the scanning system is configured to include:

a scanning platform having a matching black background plate in three dimensions;

a micro-focus camera cooperating with the background plate to take three-view shots of three dimensions of the debris;

a processing module in communication with each camera;

the processing module is used for preprocessing the image and judging the definition condition of the current shot picture through an image enhancement algorithm;

the processing module is used for carrying out primary judgment on the image definition so as to select a first position which can be used for adjusting the focal length in the image based on a judgment result;

the processing module compares and analyzes the image definition again to select a second position which can be used for adjusting the focal length on the image based on the analysis result, and the step is repeatedly carried out to obtain a corresponding lens focusing position;

the method further comprises the step of splicing and recombining the three-dimensionally scanned pictures, wherein the splicing and recombining processing method comprises the following steps:

based on the surface contour condition of the fragments obtained after scanning and the surface contour information of the fragments recorded by the scanning system, the cross sections of the fragments are compared and combined through data processing to judge whether the cross sections can be matched or not, so that the objects before impact are restored;

the method for comparing and combining the sections is configured to comprise the following steps:

taking a plane where the highest point and the lowest point of the parallel fragment section are as a base plane, taking the lowest point and the highest point of the section as base points for calculating the volume, and calculating to obtain a main volume and a missing volume of the section;

the main volume is a real volume calculated from the lowest point to the highest point of the fracture surface, the missing volume is a virtual volume calculated from the highest point to the lowest point of the fracture surface, the main volume of the fracture surface of the fragments and the missing volumes of the rest fragments are compared and analyzed to determine whether the fragments are overlapped, the matched fragment surfaces can be connected and reconstructed, and if the fragments which are not matched exist, the fragments can be subjected to dissimilation treatment and then subjected to secondary matching.

2. The method for rapidly identifying irregular particle geometries as set forth in claim 1 wherein the classification system is configured to comprise:

a control module;

the vibration grading component is arranged at a channel opening below the sputtering prevention collecting device;

the multistage conveying belt is arranged below the vibration conveying assembly;

the sorting platform is matched with the conveying belts at all levels;

the mechanical arm is arranged on one side of the sorting platform and used for conveying the classified fragments to the scanning system for three-dimensional scanning;

wherein the vibratory grading assembly is configured to include:

the device comprises a vibrating plate with the gradient larger than 10 degrees and a power mechanism matched with the vibrating plate, wherein a screening opening with multi-stage apertures is formed in the vibrating plate, and a protection plate with an arc-shaped cross section is arranged on the edge of the vibrating plate;

a first sensor disposed on the vibration plate;

each level of transmission belt is provided with a second sensor matched with the transmission belt;

and a third sensor matched with the sorting platform is arranged on the sorting platform.

3. The method for rapidly identifying irregular particle geometric features according to claim 1, wherein in the step one, the particle size classification processing mode is configured to include:

s10, triggering a first sensor when the fragments fall on the vibrating plate, wherein the first sensor transmits the acquired first signal to a control module, and the control module switches the working state of the power mechanism based on the received first signal so as to enable the vibrating plate to be in the working state;

s11, classifying the fragments falling into the vibrating plate according to the external size under the action of the continuous vibration, the gradient and the sieving port of the vibrating plate, and conveying the classified fragments to a corresponding conveying belt;

s12, triggering a second sensor through the fragments falling into the conveying belt, transmitting the acquired second signals to the control module through the second sensor, and switching the working state of the conveying belt by the control module based on the received second signals to convey the fragments of the conveying belt to the sorting platform to realize the grading treatment of the fragments.

4. The method for rapidly identifying irregular particle geometries as recited in claim 1, wherein in step five, the length characterization algorithm is configured to comprise:

on the basis of three views, an ellipsoid with three semi-major axes a, b and c in sequence is constructed, a is more than or equal to b and is more than or equal to c, assuming that in one view, a pixel lattice occupied by fragments is n, the side length of each pixel point is lambda, and on the basis of an area equivalence principle, each semi-major axis can be obtained on the basis of the following formula:

Images