Intelligent visual inspection automatic sorting system

Abstract

The invention provides an intelligent visual inspection automatic sorting system which comprises an intelligent sorting robot, the intelligent sorting robot is provided with an identification module, a walking module, a storage module and a control module, and the walking module comprises path planning. And the two-dimensional images and the three-dimensional images of the objects are identified, and the optimal path is planned according to the two-dimensional image and the three-dimensional image of each object and the sorting place, so that the walking route of the intelligent sorting robot is reduced, the possibility that the objects collide with surrounding samples is reduced, the purpose of intelligently sorting goods is achieved, and the working efficiency is improved. According to the invention, the image acquisition and processing of the object can be realized, the image recognition is accurate, and the accuracy is improved; and the problems of low manual sorting efficiency and high error rate are avoided. The object sorting efficiency is improved, the number of workers is greatly reduced, and the labor intensity is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of detection systems, and in particular to an intelligent visual detection and automatic sorting system. Background Art

[0002] In recent years, machine technologies and high-end equipment with vision as the core technology have been increasingly widely used in the intelligent manufacturing industry. This trend has greatly promoted the automation and intelligence level of the production process, not only significantly improving the traditional working environment and reducing the intensity of manual labor, but also greatly improving the manufacturing accuracy and consistency of products. Through high-precision image recognition and analysis, machine vision gives robots more precise operating capabilities, enabling them to perform more complex and sophisticated tasks on the production line.

[0003] In the existing automated sorting system, sorting robots combined with machine vision recognition technology have demonstrated powerful object recognition and processing capabilities. These robots can use advanced image recognition algorithms to quickly and accurately identify the shape, size and even material of various items, and then achieve efficient sorting and transmission through precise robot trajectory control technology. In order to adapt to the physical characteristics of different items, the robot can also flexibly adjust its clamping method, including but not limited to clamping type, supporting type and adsorption type, etc., to ensure that each item can be handled safely and stably.

[0004] However, current sorting systems often lack the ability to automatically plan the optimal walking route based on the specific sorting information of logistics items. This limitation not only limits the overall intelligence level of sorting robots, making them inflexible and inefficient when facing complex and changing sorting requirements, but also brings additional operational complexity to logistics operations, affecting the smoothness and response speed of the entire logistics chain. Therefore, there is an urgent need for an intelligent visual inspection automatic sorting system. Summary of the invention

[0005] In view of this, in order to solve the problems existing in the technical background, the present invention proposes an intelligent visual detection automatic sorting system, and the specific technical solution is as follows:

[0006] An intelligent visual detection automatic sorting system comprises an intelligent sorting robot, wherein the intelligent sorting robot is provided with a recognition module, a walking module, a storage module and a control module, wherein the walking module comprises a path planning module, the recognition module comprises a camera for collecting two-dimensional images and three-dimensional images of all sorting targets in a scene to be sorted, the recognition module is connected to the storage module and the control module, and the control module is connected to the walking module.

[0007] Furthermore, the camera is arranged on the intelligent sorting robot, and the storage module is used to store one or more programs;

[0008] Furthermore, the walking module includes obtaining a path planning image, analyzing and optimizing the path planning image, smoothing sharp corners, and setting an estimated value for each vertex (the distance from the starting point to the current vertex). Initially, only the estimated value of the starting vertex is 0, and the estimated values ​​of other vertices are infinite. At the same time, a priority queue (usually a minimum heap) is used to save the currently visited vertices, sorted from small to large by estimated value.

[0009] Take the vertex with the smallest estimated value from the priority queue, traverse all its adjacent vertices, update the estimated values ​​of these adjacent vertices (if the distance to these adjacent vertices through the current vertex is shorter), then add these adjacent vertices to the priority queue (if they are not already in the queue), and repeat the above iterative process until the target vertex is taken out of the priority queue

[0010] Furthermore, the identification module obtains basic physical properties of the object, such as weight and size.

[0011] Furthermore, the sorting area is converted into a two-dimensional grid map, and the cost evaluation function value of the node is estimated and used as the comprehensive priority of the node. When selecting the next node to be traversed, the node with the highest comprehensive priority is selected to gradually find the optimal path.

[0012] Furthermore, the walking module quickly finds the path with the smallest cost and the shortest distance.

[0013] Furthermore, the recognition module performs 3D visual recognition on the object, obtains the three-dimensional dimensions of the object on the conveyor line, and the three-dimensional spatial coordinates under the visual coordinate system, and sends the information to the storage module. When the robot is transporting, an anti-collision module is installed in the robot. The three-dimensional spatial coordinates of the object are taken into account in the path planning in the walking module, and it is judged that during this movement, there is no collision with other objects in the movement path, and then sorting and transportation are carried out.

[0014] The above technical solution has the following beneficial effects:

[0015] The present invention obtains the delivery location on the object through the recognition module, and recognizes the two-dimensional image and three-dimensional image of the object, and plans the optimal path according to the two-dimensional image and three-dimensional image of each object and the sorting location, thereby reducing the walking route of the intelligent sorting robot and reducing the possibility of collision between the object and the surrounding non-samples, thereby achieving the purpose of intelligent sorting of goods. The present invention can realize the image acquisition and processing of the object, and the image recognition is accurate, which improves the accuracy; it avoids the problems of low efficiency and high error rate of manual sorting. The present invention improves the object sorting efficiency, greatly reduces the number of workers, and reduces the labor intensity. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of the module structure of an intelligent visual inspection automatic sorting system of the present invention; DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] See also Figure 1 An intelligent visual inspection automatic sorting system shown includes an intelligent sorting robot, which is provided with a recognition module, a walking module, a storage module and a control module. The walking module includes path planning, and the recognition module includes a camera for collecting two-dimensional images and three-dimensional images of all sorting targets in the sorting scene. The recognition module is connected to the storage module and the control module, and the control module is connected to the walking module.

[0019] In this embodiment, the camera is set on the intelligent sorting robot, the storage module is used to store one or more programs, and the walking module includes obtaining the path planning image, analyzing and optimizing the path planning image, smoothing sharp corners, and setting an estimated value for each vertex (the distance from the starting point to the current vertex). Initially, only the estimated value of the starting vertex is 0, and the estimated values ​​of other vertices are infinite. At the same time, a priority queue (usually a minimum heap) is used to save the currently visited vertices, sorted from small to large by estimated value. The vertex with the smallest estimated value is taken out from the priority queue, all its adjacent vertices are traversed, and the estimated values ​​of these adjacent vertices are updated (if the distance to these adjacent vertices through the current vertex is shorter). Then, these adjacent vertices are added to the priority queue (if they are not in the queue yet), and the above iterative process is repeated until the target vertex is taken out from the priority queue. The basic physical properties of the object, such as weight and size, are obtained in the recognition module. The sorting area is converted into a two-dimensional grid map. The cost evaluation function value of the node is estimated and used as the comprehensive priority of the node. When selecting the next node to be traversed, the node with the highest comprehensive priority is selected to gradually find the optimal path. The walking module quickly finds the path with the smallest cost value and the shortest distance. The recognition module performs 3D visual recognition results on the object, obtains the three-dimensional dimensions of the length, width and height of the object on the conveyor line and the three-dimensional spatial coordinates in the visual coordinate system, and sends the information to the storage module. When the robot is transporting, an anti-collision module is provided in the robot. The three-dimensional spatial coordinates of the object are taken into account in the path planning in the walking module. It is judged that there is no collision with other objects in the motion path during this movement, and then sorting and transportation are performed.

[0020] The recognition module includes image processing to process and analyze the captured images and videos, and extract the features and attributes of the items. These features and attributes will be used for subsequent classification and path planning. In this embodiment, the control module also needs to control the sorting robot to achieve accurate grasping, placement and transportation operations to ensure the safety and accuracy of logistics items during the sorting process. In this embodiment, as the grid accuracy of the sorting area increases and the map size expands, repeated search and evaluation of useless nodes will cause the search time to grow exponentially.

[0021] The present invention obtains the delivery location on the object through the recognition module, and recognizes the two-dimensional image and three-dimensional image of the object, and plans the optimal path according to the two-dimensional image and three-dimensional image of each object and the sorting location, thereby reducing the walking route of the intelligent sorting robot and reducing the possibility of collision between the object and the surrounding non-samples, thereby achieving the purpose of intelligent sorting of goods. The present invention can realize the image acquisition and processing of the object, and the image recognition is accurate, which improves the accuracy; it avoids the problems of low efficiency and high error rate of manual sorting. The present invention improves the object sorting efficiency, greatly reduces the number of workers, and reduces the labor intensity.

[0022] The basic principles and main features of the present invention are described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for illustrating the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which shall fall within the scope of the present invention to be protected. The scope of protection of the invention shall be defined by the attached claims and their equivalents.

Claims

1. An intelligent visual inspection automatic sorting system, including an intelligent sorting robot, characterized in that: The intelligent sorting robot is provided with an identification module, a walking module, a storage module and a control module. The walking module includes path planning, the identification module includes a camera for collecting two-dimensional images and three-dimensional images of all sorting targets in the scene to be sorted, the identification module is connected to the storage module and the control module, and the control module is connected to the walking module.

2. The intelligent visual inspection automatic sorting system according to claim 1 is characterized in that: The camera is arranged on the intelligent sorting robot, and the storage module is used for storing one or more programs.

3. The intelligent visual inspection automatic sorting system according to claim 1 is characterized in that: The walking module includes obtaining the path planning image, analyzing and optimizing the path planning image, smoothing sharp corners, and setting an estimated value for each vertex. Initially, only the estimated value of the starting vertex is 0, and the estimated values ​​of other vertices are infinite. At the same time, a priority queue is used to save the currently visited vertices, sorting them from small to large by estimated value, taking out the vertex with the smallest estimated value from the priority queue, traversing all its adjacent vertices, updating the estimated values ​​of these adjacent vertices, and then adding these adjacent vertices to the priority queue. Repeat the above iterative process until the target vertex is taken out of the priority queue.

4. The intelligent visual inspection automatic sorting system according to claim 1 is characterized in that: The recognition module obtains basic physical properties of the object, such as weight and size.

5. The intelligent visual inspection automatic sorting system according to claim 3 is characterized in that: The sorting area is converted into a two-dimensional grid map. The cost evaluation function value of the node is estimated and used as the comprehensive priority of the node. When selecting the next node to be traversed, the node with the highest comprehensive priority is selected to gradually find the optimal path.

6. The intelligent visual inspection automatic sorting system according to claim 1 is characterized in that: The walking module quickly finds the path with the minimum cost and the shortest distance.

7. The intelligent visual inspection automatic sorting system according to claim 2 is characterized in that: The recognition module performs 3D visual recognition on the object, obtains the three-dimensional dimensions of the length, width and height of the object on the conveyor line and the three-dimensional spatial coordinates under the visual coordinate system, and sends the information to the storage module. When the robot is transporting, an anti-collision module is installed in the robot. The three-dimensional spatial coordinates of the object are taken into account in the path planning in the walking module. It is determined that there is no collision with other objects in the movement path during this movement, and then sorting and transportation are carried out.

Images