A warehouse logistics detection system based on visual recognition

Abstract

The application discloses a kind of warehouse logistics detection systems based on visual identification, it is related to logistics storage technical field, solve the technical problems that current technique easily causes goods information omission and goods loss, lead to the technical problem that storage efficiency is not high, cannot guarantee storage quality;The present application comprises central analysis module, and is connected with several image acquisition equipment;Several image acquisition equipment is arranged in several storage functional areas of storage area, and storage functional area includes logistics warehousing area, logistics storage area and logistics distribution area;The image acquisition equipment in the present application is obtained in the logistics warehousing area, and the label feature and appearance feature of logistics package are identified according to the warehousing image;In logistics storage area and logistics distribution area, logistics package is identified by global image, and then matching record storage location and distribution information, realize the automation of storage logistics;The present application can realize the automatic detection of storage logistics, improve storage logistics efficiency, reduce labor cost.

Description

Technical Field

[0001] This invention belongs to the field of logistics and warehousing, and relates to warehousing and logistics inspection technology based on visual recognition, specifically a warehousing and logistics inspection system based on visual recognition. Background Technology

[0002] Warehousing and logistics involves storing, managing, and distributing goods using self-built or leased warehouses and facilities. With the development of the express delivery industry, the scale of the warehousing and logistics industry has also grown, making the tracking and location of goods in complex warehousing environments increasingly important.

[0003] Current technologies typically use handheld devices or smartphones to record the storage, safekeeping, and distribution of goods, requiring manual handling for each item. However, these existing technologies are prone to errors during warehouse management, leading to omissions of goods information and frequent instances of lost goods. Therefore, current technologies impact warehousing efficiency and cannot guarantee storage quality. Consequently, there is an urgent need for a vision-based warehouse logistics inspection system. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a warehousing and logistics detection system based on visual recognition to solve the technical problem that the prior art is prone to omission of cargo information and loss of cargo, resulting in low warehousing efficiency and inability to guarantee warehousing quality.

[0005] To achieve the above objectives, a first aspect of the present invention provides a visual recognition-based warehousing and logistics inspection system, including a central analysis module and several image acquisition devices connected thereto; the several image acquisition devices are installed in several warehousing functional areas of the warehousing area, including a logistics receiving area, a logistics storage area and a logistics distribution area. The central analysis module uses image acquisition devices in the logistics warehousing area to capture images of logistics packages entering the warehouse in an orderly manner and marks them as warehousing images; the warehousing images include the logistics tags of the logistics packages; The central analysis module determines the label and appearance features of the logistics packages based on the images of the packages entering the warehouse and establishes a correlation between the two; then, it acquires stored images of the logistics packages through the image acquisition equipment in the logistics storage area. The central analysis module analyzes whether the logistics packages in the stored images match the appearance features; if yes, it extracts the storage location from the stored images and associates the storage location with the tag features; if no, it issues an alert. The central analysis module acquires delivery images of logistics packages through image acquisition equipment in the logistics distribution area, and analyzes whether the logistics packages in the delivery images match the appearance features; if yes, the delivery information is extracted and associated with the label features; otherwise, the detection continues.

[0006] Preferably, the central analysis module communicates with and / or connects to a smart terminal and several image acquisition devices; wherein the smart terminal includes a mobile phone or a computer; Several of the aforementioned image acquisition devices are rationally arranged in the storage functional area; and the logistics warehousing area, logistics storage area and logistics distribution area are connected by intelligent vehicles.

[0007] Preferably, the central analysis module collects images of incoming logistics packages in the logistics receiving area, including: After adjusting the posture and spacing of the logistics packages, they are transported in an orderly manner on the package conveyor belt in the logistics warehouse area. Several image acquisition devices in the logistics warehousing area sequentially acquire images of logistics packages entering the warehouse. After preprocessing the images, they are sent to the central analysis module. The images are captured from different angles.

[0008] Preferably, the central analysis module obtains the label features and appearance features of the logistics package based on the image recognition of the incoming warehouse, including: Select at least one inbound image with a logistics tag, and obtain the tag features through the identification of the inbound image; at the same time, construct a three-dimensional outline of the logistics package based on several inbound images. The size characteristics of the logistics package are obtained based on the three-dimensional contour, and the appearance characteristics are generated by combining the color characteristics of the logistics package. After associating the label characteristics and appearance characteristics, the logistics package is transported by a smart vehicle.

[0009] Preferably, the central analysis module obtains the size characteristics of the logistics package based on the three-dimensional contour, including: The length and width of at least three asymmetric planes of a logistics package are obtained based on the three-dimensional contour. The lengths of the three asymmetric planes are labeled as MC1, MC2 and MC3, and the widths are labeled as MK1, MK2 and MK3, respectively, generating the dimension feature as [(MC1, MK1), (MC2, MK2), (MC3, MK3)].

[0010] Preferably, the central analysis module obtains the color features of the logistics package through several inbound images, including: Identify three asymmetric planes and recognize the feature points of the three asymmetric planes; Calculate and obtain the position data of feature points on the corresponding symmetry plane; The location data is combined with the size features to generate the appearance features.

[0011] Preferably, the central analysis module performs storage detection based on the stored images of logistics packages, including: Warehouse images of the logistics storage area are acquired using image acquisition equipment, and storage images of logistics packages are extracted based on warehouse image recognition; the warehouse images include several logistics packages. If the features of the stored image match at least one set of data in the appearance features, then the storage location of the logistics package is extracted, and the association between the storage location and the corresponding label features of the logistics package is established.

[0012] Preferably, the central analysis module performs delivery detection based on delivery images of logistics parcels, including: Images of the logistics and distribution area are captured using image acquisition equipment; delivery images of logistics packages are extracted based on the images of the logistics and distribution area; the images of the logistics and distribution area include several logistics packages. If the features of the delivery image match at least one set of data in the appearance features, then the delivery information is extracted and the association between the delivery information and the corresponding label features of the logistics package is established.

[0013] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention acquires images of the incoming logistics through an image acquisition device set in the logistics receiving area, and identifies the label features and appearance features of the logistics packages based on the images; in the logistics storage area and logistics distribution area, the logistics packages are identified through global images, and then the storage location and delivery information are matched and recorded to realize the automation of warehousing and logistics; the present invention can realize the automated detection of warehousing and logistics, improve the efficiency of warehousing and logistics, and reduce labor costs. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a schematic diagram illustrating the working steps of the present invention. Detailed Implementation

[0016] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Please see Figure 1 The first aspect of this invention provides a visual recognition-based warehousing and logistics detection system, including a central analysis module and several image acquisition devices connected thereto. The image acquisition devices are arranged in several functional areas of a warehousing area, including a logistics entry area, a logistics storage area, and a logistics distribution area. The central analysis module acquires images of orderly entering logistics packages through the image acquisition devices in the logistics entry area, marking them as entry images. The central analysis module determines the label features and appearance features of the logistics packages based on the entry images and establishes a correlation between them. Then, it acquires storage images of the logistics packages through the image acquisition devices in the logistics storage area. The central analysis module analyzes whether the logistics packages in the storage images match the appearance features; if yes, it extracts the storage location from the storage image and associates the storage location with the label features; if no, it issues an early warning. The central analysis module acquires delivery images of the logistics packages through the image acquisition devices in the logistics distribution area and analyzes whether the logistics packages in the delivery images match the appearance features; if yes, it extracts delivery information and associates the delivery information with the label features; if no, it continues detection.

[0018] Current warehousing and logistics inspection technologies mostly rely on handheld devices or smartphones to scan the logistics tags on packages. After identifying the information, packages are manually stored in designated areas and retrieved from storage locations based on delivery information during delivery. This process requires constant manual intervention from staff, which is not only labor-intensive but also inefficient. Furthermore, staff errors can compromise storage quality.

[0019] In this invention, the central analysis module communicates with and / or connects to a smart terminal and several image acquisition devices; wherein the smart terminal includes a mobile phone or a computer; the several image acquisition devices are reasonably arranged in the storage functional area; and the logistics warehousing area, logistics storage area and logistics distribution area are connected by smart carts.

[0020] The central analysis module is the core control component of the entire warehousing and logistics process. It primarily analyzes the collected image data to establish the entire warehousing process for logistics packages, mainly interacting with various image acquisition devices. These devices, primarily cameras, are distributed across various storage functional areas, with each area containing more than one camera. The storage functional areas are arranged according to the order of package detection: the logistics receiving area, the logistics storage area, and the logistics distribution area. Intelligent carts transport packages between these areas.

[0021] In a preferred embodiment, the central analysis module collects images of logistics packages entering the warehouse in the logistics warehousing area, including: adjusting the posture and spacing of the logistics packages and then conveying them in an orderly manner on the package conveyor belt in the logistics warehousing area; several image acquisition devices in the logistics warehousing area sequentially collect images of the logistics packages entering the warehouse, preprocess the images, and then send them to the central analysis module.

[0022] Before capturing images of the incoming packages, it's crucial to ensure the image acquisition equipment can capture the shipping labels on the parcels. This means avoiding positioning the labels towards the conveyor belt while maintaining proper spacing between the parcels. Once the parcels enter the working area of ​​the image acquisition equipment in an orderly manner, the equipment will capture the incoming images. It's important to note that different image acquisition devices will capture images from different angles, resulting in multiple incoming images. These multiple images must be sufficient to establish a three-dimensional outline.

[0023] In a preferred embodiment, the central analysis module obtains the label features and appearance features of the logistics package based on the warehouse entry image recognition, including: selecting at least one warehouse entry image carrying a logistics label, and obtaining the label features through the warehouse entry image recognition; simultaneously, establishing a three-dimensional outline of the logistics package based on several warehouse entry images; obtaining the size features of the logistics package based on the three-dimensional outline, and generating appearance features by combining the color features of the logistics package; after associating the label features and appearance features, transporting the logistics package through an intelligent vehicle.

[0024] The process begins by identifying logistics tags from selected inbound images, and then extracting logistics information, or tag features, based on these tags. Next, size features are identified from several inbound images, leading to the generation of appearance features. Size features are primarily used to identify the approximate shape of the package. Regular shapes, such as cuboids, are easily identified; for irregular shapes, their shape features can be determined from the inbound images.

[0025] In an optional embodiment, the central analysis module obtains the size features of the logistics package based on the three-dimensional contour, including: obtaining the length and width of at least three asymmetric surfaces of the logistics package based on the three-dimensional contour; marking the lengths of the three asymmetric surfaces as MC1, MC2 and MC3 respectively, and marking the widths as MK1, MK2 and MK3 respectively, generating the size features as [(MC1, MK1), (MC2, MK2), (MC3, MK3)].

[0026] For regular shapes, the length and width of their three asymmetric faces can be calculated, and then dimensional features can be established based on the length and width of each face. Irregular shapes are less common, and very few are exactly the same in the same warehouse; therefore, it is only necessary to determine their shape and the necessary dimensional data.

[0027] In an optional embodiment, the central analysis module obtains the color features of the logistics package through several inbound images, including: determining three asymmetric surfaces, identifying feature points on the three asymmetric surfaces; calculating the position data of the feature points on the corresponding symmetric surfaces, and combining the position data with the size features to generate appearance features.

[0028] After determining the size characteristics, it is not possible to completely distinguish logistics packages based solely on size characteristics; therefore, color characteristics are needed. This embodiment identifies the color of each surface. If a special mark appears, such as a red mark on a yellow cardboard box, the red mark is used as a feature, i.e., a feature point. The position of the red mark on the corresponding surface is calculated as the color feature. If a surface does not have a special mark, its grayscale mean is calculated after preprocessing as the color feature.

[0029] In a preferred embodiment, the central analysis module performs storage detection based on the storage image of the logistics package, including: acquiring warehouse images of the logistics storage area through an image acquisition device, and extracting the storage image of the logistics package based on the warehouse image recognition; identifying and analyzing whether the features of the storage image match at least one set of data in the appearance features; if so, extracting the storage location of the logistics package and establishing the association between the storage location and the corresponding label features of the logistics package.

[0030] The intelligent vehicle transports logistics packages from the inbound warehouse area to the logistics storage area. At this stage, neither staff nor the intelligent vehicle need to collect information from the packages. The image acquisition device in the logistics storage area acquires a global image, i.e., a stored image, and identifies several logistics packages within the stored image, along with their corresponding appearance features. Due to the storage environment, the identified appearance features may only include the size and color features of one face. These face's size and color features are compared with the appearance features of each logistics package. Only when a set of appearance features matches the size and color features of that face can the logistics package corresponding to that face be determined, its storage location extracted, and associated with the corresponding package's tag features. This embodiment can quickly locate logistics packages without manual processing.

[0031] In a preferred embodiment, the central analysis module performs delivery detection based on delivery images of logistics parcels, including: acquiring working images of the logistics delivery area through an image acquisition device; identifying and extracting delivery images of logistics parcels based on the working images; identifying and analyzing whether the features of the delivery images match at least one set of data in the appearance features; if so, extracting delivery information and establishing a correlation between the delivery information and the corresponding label features of the logistics parcels.

[0032] This embodiment identifies logistics packages in the working image, extracts the corresponding delivery image, and then determines which logistics package is being delivered based on the delivery image. The delivery information is then associated with the corresponding logistics package's tag features. Alternatively, after obtaining the delivery information, the logistics package can be transported to the logistics distribution area based on previously obtained storage locations. No manual intervention is required during the delivery process to complete the recording of delivery information, reducing costs.

[0033] The warning in this invention specifically refers to a situation where a logistics package is stored in the storage area, but the corresponding label features and appearance features of the logistics package are not found in the system. This may be due to a processing error by the central analysis module, and therefore a warning needs to be issued.

[0034] Working principle of the invention: The central analysis module acquires images of orderly entering logistics packages through image acquisition devices in the logistics warehousing area and marks them as warehousing images. The central analysis module determines the label features and appearance features of the logistics packages based on the warehousing images and establishes the correlation between the two. Then, it acquires storage images of the logistics packages through image acquisition devices in the logistics storage area.

[0035] The central analysis module analyzes whether the logistics packages in the stored images match the appearance features; if yes, it extracts the storage location from the stored images and associates the storage location with the tag features; if no, it issues an alert.

[0036] The central analysis module acquires delivery images of logistics packages through image acquisition equipment in the logistics distribution area, and analyzes whether the logistics packages in the delivery images match the appearance features; if yes, the delivery information is extracted and associated with the label features; otherwise, the detection continues.

[0037] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A vision-based warehousing and logistics inspection system, comprising a central analysis module and several image acquisition devices connected thereto; the image acquisition devices are installed in several functional areas of the warehousing area, including a logistics receiving area, a logistics storage area, and a logistics distribution area; characterized in that: The central analysis module collects images of incoming logistics packages in the logistics receiving area, including: After adjusting the posture and spacing of the logistics packages, they are transported in an orderly manner on the package conveyor belt in the logistics warehouse area. Several image acquisition devices in the logistics warehousing area sequentially acquire images of logistics packages entering the warehouse. After preprocessing the images, they are sent to the central analysis module. The images are taken from different angles and include the logistics labels on the packages. The central analysis module obtains the label features and appearance features of logistics packages based on inbound image recognition, and establishes the correlation between the two, including: Select at least one inbound image with a logistics tag, and obtain the tag features through inbound image recognition; at the same time, construct a three-dimensional outline of the logistics package based on several inbound images. The size features of the logistics package are obtained based on the three-dimensional contour, and the appearance features are generated by combining the color features of the logistics package; after associating the label features and appearance features, the logistics package is transported by a smart vehicle. The central analysis module obtains the dimensional features of logistics packages based on three-dimensional contours, including: The length and width of at least three asymmetric planes of a logistics package are obtained based on the three-dimensional contour. The lengths of the three asymmetric planes are labeled as MC1, MC2 and MC3, and the widths are labeled as MK1, MK2 and MK3, respectively, generating the dimension feature [(MC1, MK1), (MC2, MK2), (MC3, MK3)]; The central analysis module obtains the color characteristics of logistics packages through several inbound images, including: Identify three asymmetric planes and recognize the feature points of the three asymmetric planes; Calculate and obtain the position data of feature points on the corresponding symmetry plane; Combine location data with size features to generate appearance features; Next, the image acquisition device in the logistics storage area acquires storage images of the logistics packages; the central analysis module analyzes whether the logistics packages in the storage images match the appearance features; if yes, the storage location is extracted from the storage image and associated with the tag features; if no, an early warning is issued. The central analysis module acquires delivery images of logistics packages through image acquisition equipment in the logistics distribution area, and analyzes whether the logistics packages in the delivery images match the appearance features; if yes, the delivery information is extracted and associated with the label features; if no, the detection continues. The logistics receiving area, logistics storage area and logistics distribution area are connected by intelligent vehicles.

2. The warehousing and logistics inspection system based on visual recognition according to claim 1, characterized in that, The central analysis module communicates with and / or connects to smart terminals and several image acquisition devices; wherein, the smart terminals include mobile phones or computers; Several of the aforementioned image acquisition devices are rationally arranged in the warehouse functional area.

3. The warehousing and logistics inspection system based on visual recognition according to claim 1, characterized in that, The central analysis module performs storage detection based on the stored images of logistics packages, including: Warehouse images of the logistics storage area are acquired using image acquisition equipment, and storage images of logistics packages are extracted based on warehouse image recognition; the warehouse images include several logistics packages. If the features of the stored image match at least one set of data in the appearance features, then the storage location of the logistics package is extracted, and the association between the storage location and the corresponding label features of the logistics package is established.

4. The warehousing and logistics inspection system based on visual recognition according to claim 1, characterized in that, The central analysis module performs delivery detection based on delivery images of logistics parcels, including: Images of the logistics and distribution area are captured using image acquisition equipment; delivery images of logistics packages are extracted based on the images of the logistics and distribution area; the images of the logistics and distribution area include several logistics packages. If the features of the delivery image match at least one set of data in the appearance features, then the delivery information is extracted and the association between the delivery information and the corresponding label features of the logistics package is established.

Images