Visual algorithm-based bulk cargo wharf storage yard intelligent management system

Abstract

The invention provides a general cargo wharf storage yard intelligent management system based on a visual algorithm. The system comprises a database, an image acquisition module, a visual algorithm module and an intelligent management module. The database module stores management data such as contours, sizes, colors, types and positions. The image acquisition module acquires a storage yard image in real time. And the visual algorithm module determines cargo types by matching contour, size and color data, compares the types of data to obtain identification accuracy, and performs repeated processing if the identification accuracy does not reach the standard until the identification accuracy reaches the standard. The intelligent management module counts position data, compares the position data with position parameters to obtain position indexes, and triggers alarm when the position indexes exceed standards. The system integrates all functional modules, and realizes intelligent management including data storage, real-time image acquisition, cargo information determination through image analysis, statistical analysis and exception handling.

Description

Technical Field

[0001] The present application belongs to the field of piece goods management at a terminal, and in particular, relates to an intelligent management system for piece goods terminal yards based on a visual algorithm. Background Art

[0002] Breakbulk cargo, as a general term for goods that are counted by piece, of various shapes and sizes, has always been a complex and inefficient process for transportation and yard management. This type of cargo cannot be transported in a standardized manner using containers, so the loading and unloading process is complex and requires frequent human-machine interaction. In breakbulk cargo terminal yards, there have long been problems such as low operating efficiency and time-consuming management. Traditional terminal yard management mainly relies on manual operation and simple mechanized equipment, which not only consumes a lot of time and energy, but is also easily affected by human factors, resulting in low management efficiency and inaccurate cargo identification. In addition, due to the limitations of manual operation, yard space resources cannot be fully utilized, and cargo is often stacked in a chaotic and disorderly manner, further exacerbating the difficulty of management.

[0003] With the rapid development of machine vision technology, its application in the field of logistics is becoming more and more extensive, especially in the identification, counting and monitoring of goods. However, in the field of yard management technology, due to the instability of the types, packaging, dimensions and placement of piece goods, the management of piece goods is much more difficult than that of boxed goods. Traditional image analysis or artificial intelligence algorithms often find it difficult to accurately and effectively identify, locate and monitor such complex and changeable goods. Summary of the invention

[0004] The purpose of this application is to overcome the defects in the above-mentioned prior art and provide a general cargo terminal yard intelligent management system based on visual algorithm.

[0005] The present application provides a general cargo terminal yard intelligent management system based on a visual algorithm, including: A database module for storing management data of the general cargo on the terminal yard, wherein the management data includes contour parameters, size parameters, color parameters, type parameters and location parameters; An image acquisition module for acquiring image data of the piece goods on the terminal yard in real time; Visual algorithm module, S101, obtaining the position data, outline data, size data and color data of the piece of groceries; S102, matching the outline data, size data and color data with the outline parameters, size parameters and color parameters to determine the type data of the piece of groceries; S103, comparing the type data with the type parameters, and obtaining a recognition accuracy index by statistically comparing the results; S104, if the recognition accuracy index is less than a first threshold, repeating S101 to S104, and if the recognition accuracy index is not less than the first threshold, entering the intelligent management module; The intelligent management module collects statistics on the location data of the different types of groceries to obtain location coordinate data; compares the location coordinate data with the location parameters, and obtains a location index by collecting statistics of the comparison results; and triggers an alarm mechanism if the location index is greater than a second threshold.

[0006] Optionally, it also includes: The intelligent monitoring module obtains a stability monitoring index according to the ground area, height and quantity of the piece of groceries; triggers an alarm mechanism if the stability monitoring index is greater than a third threshold; determines a tracking monitoring index according to the quantity, and triggers an alarm mechanism if the tracking monitoring index is greater than a fourth threshold.

[0007] Optionally, the image acquisition module includes a high-definition camera.

[0008] Optionally, when acquiring the position data, contour data, size data and color data of the piece goods, the visual algorithm module first pre-processes the image data by using image processing technology.

[0009] Optionally, the preprocessing includes image resizing, color correction, denoising and image enhancement.

[0010] Optionally, the recognition accuracy index is obtained through big data analysis technology, including the ratio of the number of correctly recognized piece goods to the total number of recognized pieces.

[0011] Optionally, when the alarm mechanism is triggered, the intelligent management module also sends the alarm information to the human-computer interaction module, so that the administrator can handle the abnormal situation in time.

[0012] Optionally, the factors for calculating the stability monitoring index also include: a stacking inclination angle.

[0013] Optionally, the intelligent management module includes a position calibration unit, which is used to compare the position coordinates of the piece goods obtained by the visual algorithm with the position information in the database to calibrate the actual position of the piece goods.

[0014] Optionally, the intelligent monitoring module includes a historical data comparison unit, which is used to compare the currently monitored cargo stacking status with historical data and analyze the changing trend of cargo stacking.

[0015] The beneficial effects of this application are: The present application provides an intelligent management system for piece goods terminal yard based on visual algorithms, comprising: a database module, storing management data of piece goods on the terminal yard, wherein the management data comprises contour parameters, size parameters, color parameters, type parameters and position parameters; an image acquisition module, collecting image data of the piece goods on the terminal yard in real time; a visual algorithm module, S101, acquiring position data, contour data, size data and color data of the piece goods; S102, matching the contour data, size data and color data with the contour parameters, size parameters and color parameters to determine the position of the piece goods. The type data of the groceries; S103, compare the type data with the type parameter, and obtain the recognition accuracy index by statistical comparison results; S104, if the recognition accuracy index is less than the first threshold, repeat S101~S104, if the recognition accuracy index is not less than the first threshold, enter the intelligent management module; the intelligent management module counts the position data of the different types of the groceries to obtain the position coordinate data; compare the position coordinate data with the position parameter, and obtain the position index by statistical comparison results; if the position index is greater than the second threshold, trigger the alarm mechanism. By integrating multiple functional modules such as database module, image acquisition module, visual algorithm module and intelligent management module, the intelligent management of the groceries terminal yard is realized. Among them, the database module is used to store and manage various parameter information of groceries; the image acquisition module is used to collect image data of goods on the yard in real time; the visual algorithm module is responsible for processing and analyzing image data to determine the type and location of goods and other information; the intelligent management module performs statistics and analysis based on the results of the visual algorithm module to timely discover and handle abnormal situations. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic diagram of the intelligent management system for general cargo terminal yard based on visual algorithm in this application. DETAILED DESCRIPTION

[0017] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, the embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0018] Please refer to Figure 1As shown, the present application relates to an intelligent management system for a general cargo terminal yard based on a visual algorithm, including: A database module for storing management data of the general cargo on the terminal yard, wherein the management data includes contour parameters, size parameters, color parameters, type parameters and location parameters; Through big data technology, the management data of general cargo at the terminal yard is stored, and a general cargo terminal yard database is built.

[0019] The management data of piece goods includes but is not limited to the type parameters, size parameters, outline parameters, color parameters, position parameters, etc. of the piece goods. The database module is connected to all modules in the system to receive and store the data of each module and update it in real time.

[0020] An image acquisition module for acquiring image data of the piece goods on the terminal yard in real time; The image acquisition module collects images of the general cargo on the dock yard in real time through the image acquisition device, obtains the image data of the general cargo to be managed, and transmits the image data of the general cargo to be managed to the visual algorithm module; The collected image data has been pre-processed through image processing technology, such as adjusting image size, color correction, denoising, etc., to improve image quality; image enhancement, such as contrast enhancement, brightness adjustment, etc., has been performed to highlight the characteristics of the goods.

[0021] Visual algorithm module, S101, obtaining the position data, outline data, size data and color data of the piece of groceries; S102, matching the outline data, size data and color data with the outline parameters, size parameters and color parameters to determine the type data of the piece of groceries; S103, comparing the type data with the type parameters, and obtaining a recognition accuracy index by statistically comparing the results; S104, if the recognition accuracy index is less than a first threshold, repeating S101 to S104, and if the recognition accuracy index is not less than the first threshold, entering the intelligent management module; The visual algorithm module performs visual algorithm processing on the image data of the piece goods to be managed obtained by the image acquisition module, and transmits the visual algorithm processing results of the piece goods to the intelligent management module. The visual algorithm module includes: a visual algorithm processing unit, a visual algorithm recognition unit and a visual algorithm processing evaluation unit.

[0022] Vision algorithm processing unit: Through the image data of the piece goods, the visual algorithm is used to obtain the contour data CD of the piece goods, the size data of the piece goods, including the three-dimensional size data SD and the color data CLD of the piece goods, respectively. The expression is: ; Where n represents the number of pieces of groceries, represents the outline of the i-th piece of groceries, represents the three-dimensional size of the i-th piece of groceries, Indicates the color of the i-th piece of grocery.

[0023] The outline data is obtained by identifying the piece goods through image processing technology and connecting the edge points to form a closed outline. The three-dimensional size of the piece goods refers to the data on the x, y and z axes with the center of the piece goods as the origin. The color of the piece goods refers to the values ​​of the three primary colors of red R, green G and blue B of the piece goods.

[0024] Visual algorithm recognition unit: Obtain the recognition data of each piece of groceries based on the contour data CD, the three-dimensional size data SD and the color data CLD of the piece of groceries : ; Then, the identification data of each piece of general cargo is traversed and matched with the cargo information in the general cargo terminal yard database. The cargo information includes contour parameters, size parameters and color parameters, and the type matching index data of each piece of general cargo is obtained. : ; Among them, k represents the number of category matching indexes, represents the jth category matching index, classifies the i-th piece goods as the category with the largest category matching index, and obtains the category identification data TD of the piece goods: ; Among them, m represents the number of types of groceries, Indicates the first type of general cargo, for example, the types of general cargo identified are steel and wood, and the matching index MC is: ; Wherein, η(CD) represents the contour matching coefficient of the piece goods, η(CD)=Sove / S, Sove represents the overlapping area between the contour of the piece goods and the contour of the goods in the database, and S represents the contour area of ​​the goods in the database.

[0025] ; and (x, y, z) represent the three-dimensional size data of the piece goods and the three-dimensional size of the goods in the database respectively.

[0026] ; and (x, y, z) represent the red, green and blue primary color values ​​of the groceries and the red, green and blue primary color values ​​of the goods in the database respectively, and a1, a2 and a3 represent the corresponding weights respectively, for example, a1=0.4, a2=0.3 and a3=0.3.

[0027] Based on the category identification data TD, the category quantity data Q(TD) of the piece goods is obtained by statistics: ; The quantity of the first type of groceries is .

[0028] Visual algorithm processing evaluation unit: Based on the type and quantity data of piece goods obtained by visual algorithm processing, the visual algorithm processing evaluation index PEI is obtained through big data analysis technology: ; in, and They represent the number of correctly identified piece goods contours and the total number of piece goods contours, and They represent the number of correctly identified piece goods types and the total number of piece goods types respectively, and b1 and b2 represent the corresponding weights respectively, for example, b1=0.5 and b2=0.5.

[0029] If the PEI is less than the preset first threshold, the visual algorithm processing unit and the visual algorithm recognition unit are repeated, otherwise the intelligent management module is entered.

[0030] The intelligent management module collects statistics on the location data of the different types of groceries to obtain location coordinate data; compares the location coordinate data with the location parameters, and obtains a location index by collecting statistics of the comparison results; and triggers an alarm mechanism if the location index is greater than a second threshold.

[0031] First, a general cargo terminal yard coordinate system is established through a visual algorithm to identify the positions of different types of general cargo in real time, locate the position of the general cargo on the yard, and obtain the position coordinate data PCD of the general cargo on the yard: ; Where K represents the number of different position coordinates, Indicates the location coordinates of the Jth piece of general cargo in the yard.

[0032] Through big data analysis technology, we can obtain the location identification management indicators for each piece of groceries: ; in, represents (x,y) and Respectively represent the coordinates of each piece of general cargo located by the visual algorithm and the coordinates of the corresponding cargo in the general cargo terminal yard database. If it is greater than the preset second threshold, it means that the placement of the groceries is abnormal, triggering the alarm mechanism, and automatically sending an alarm message to the human-computer interaction module. Otherwise, it is normal.

[0033] The intelligent monitoring module obtains a stability monitoring index according to the ground area, height and quantity of the piece of groceries; triggers an alarm mechanism if the stability monitoring index is greater than a third threshold; determines a tracking monitoring index according to the quantity, and triggers an alarm mechanism if the tracking monitoring index is greater than a fourth threshold.

[0034] The piece goods are monitored according to their location coordinates on the yard. The alarm mechanism is automatically triggered according to the abnormal monitoring results, and the alarm information is transmitted to the human-computer interaction module. The intelligent monitoring module includes a cargo stacking stability risk monitoring unit and a cargo stacking tracking risk monitoring unit: Cargo stacking stability risk monitoring unit: The visual algorithm is used to obtain the ground area A occupied by each piece of general cargo in the yard, the height h of each piece of general cargo, and the quantity QG of each piece of general cargo, and the cargo stacking stability risk monitoring index SMI is obtained: ; Wherein, θ represents the inclination angle of the piece goods stacking. If the SMI is greater than the preset third threshold, it means that the stability of the goods stacking is abnormal, and the alarm mechanism is automatically triggered, and the alarm information is transmitted to the human-computer interaction module.

[0035] Cargo stacking tracking risk monitoring unit: Through visual algorithms, the quantity QG of each type of piece goods on the yard is monitored in real time to track the in and out information of the piece goods and obtain the cargo stacking tracking risk monitoring index EMI: ; in, represents the quantity of corresponding goods in the general cargo terminal yard database, and γ represents the continuity coefficient of cargo movement tracking: ; in, Indicates the number of frames included in the piece goods tracking process. It represents the number of frames lost in tracking, and d represents the distance of the cargo position jump between adjacent frames.

[0036] If the EMI is greater than the preset fourth threshold, it means that the monitoring detects abnormal risk of cargo stacking tracking, automatically triggers the alarm mechanism, and transmits the alarm information to the human-computer interaction module.

[0037] Human-computer interaction module: used to receive alarm information from the intelligent monitoring module and conduct human-computer interaction, prompting the manager of the general cargo terminal to take timely measures. For example, when risks of cargo stacking stability are found, measures such as strengthening the stacking structure and reducing the stacking height are taken; for example, when risks of cargo stacking tracking are found, the cargo is immediately verified to ensure its safety.

[0038] Preferably, the image acquisition module includes a high-definition camera.

[0039] Preferably, the factors for calculating the stability monitoring index also include: stacking inclination angle.

[0040] Preferably, the intelligent management module includes a position calibration unit for comparing the position coordinates of the piece goods obtained by the visual algorithm with the position information in the database to calibrate the actual position of the piece goods.

[0041] Preferably, the intelligent monitoring module includes a historical data comparison unit for comparing the currently monitored cargo stacking status with historical data to analyze the changing trend of cargo stacking.

[0042] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent management system for general cargo terminal yard based on visual algorithm, characterized in that: include: A database module for storing management data of the general cargo on the terminal yard, wherein the management data includes contour parameters, size parameters, color parameters, type parameters and location parameters; An image acquisition module for acquiring image data of the piece goods on the terminal yard in real time; Visual algorithm module, S101, obtaining the position data, outline data, size data and color data of the piece of groceries; S102, matching the outline data, size data and color data with the outline parameters, size parameters and color parameters to determine the type data of the piece of groceries; S103, comparing the type data with the type parameters, and obtaining a recognition accuracy index by statistically comparing the results; S104, if the recognition accuracy index is less than a first threshold, repeating S101 to S104, and if the recognition accuracy index is not less than the first threshold, entering the intelligent management module; The intelligent management module collects statistics on the location data of the different types of groceries to obtain location coordinate data; compares the location coordinate data with the location parameters, and obtains a location index by collecting statistics of the comparison results; and triggers an alarm mechanism if the location index is greater than a second threshold.

2. According to claim 1, a visual algorithm-based intelligent management system for general cargo terminal yards, characterized in that: Also includes: An intelligent monitoring module obtains stability monitoring indicators according to the ground area, height and quantity of the piece of groceries; If the stability monitoring index is greater than the third threshold, an alarm mechanism is triggered; based on the quantity, a tracking monitoring index is determined, and if the tracking monitoring index is greater than the fourth threshold, an alarm mechanism is triggered.

3. The intelligent management system for general cargo terminal yard based on visual algorithm according to claim 1 is characterized in that: The image acquisition module includes a high-definition camera.

4. The intelligent management system for general cargo terminal yard based on visual algorithm according to claim 1 is characterized in that: When acquiring the position data, contour data, size data and color data of the piece goods, the visual algorithm module first pre-processes the image data by using image processing technology.

5. The intelligent management system for general cargo terminal yard based on visual algorithm according to claim 4 is characterized in that: The preprocessing includes image resizing, color correction, denoising, and image enhancement.

6. The intelligent management system for general cargo terminal yard based on visual algorithm according to claim 1 is characterized in that: The recognition accuracy index is obtained through big data analysis technology, including the ratio of the number of correctly recognized piece goods to the total number of recognized pieces.

7. The intelligent management system for general cargo terminal yard based on visual algorithm according to claim 1 is characterized in that: When the alarm mechanism is triggered, the intelligent management module also sends the alarm information to the human-computer interaction module so that the administrator can handle the abnormal situation in time.

8. The intelligent management system for general cargo terminal yard based on visual algorithm according to claim 2 is characterized in that: Factors for calculating the stability monitoring index also include: stacking inclination angle.

9. The intelligent management system for general cargo terminal yard based on visual algorithm according to claim 1 is characterized in that: The intelligent management module includes a position calibration unit, which is used to compare the position coordinates of the piece goods obtained by the visual algorithm with the position information in the database to calibrate the actual position of the piece goods.

10. The intelligent management system for general cargo terminal yard based on visual algorithm according to claim 2 is characterized in that: The intelligent monitoring module includes a historical data comparison unit, which is used to compare the currently monitored cargo stacking status with historical data and analyze the changing trend of cargo stacking.

Images