Logistics express sheet identification method, system and device
An identification method and identification device technology, applied in the field of logistics identification, can solve problems such as slow processing speed and access costs, and achieve the effects of improving processing speed and reducing costs.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0037] Such as figure 1 As shown, a logistics bill identification method includes the following steps:
[0038] S1. Photograph the upward surface of the express package to obtain an image;
[0039] S2. Segment the image and extract the face sheet;
[0040] S3. Perform rotation correction on the extracted face sheet, and rotate 180 degrees if the face sheet is reversed;
[0041] S4. Detect the receiving address and sorting code on the face sheet;
[0042] S5. Identify the detected receiving address and sorting code;
[0043] S6. Match the identified receiving address and sorting code with the data in the database to obtain the receiving site; if the detection result does not match the data in the database, it means that the detection is unsuccessful and enters the recycling area;
[0044] S7. Generate a sorting instruction and send it to the sorting device;
[0045] S8. The sorting device sorts and transports the package to the delivery port that matches the receiving site...
Embodiment 2
[0060] Such as figure 2 As shown, a logistics bill recognition system includes: a lighting module, an image acquisition module, a face bill extraction module, a receiving address and sorting code extraction module, and an address matching module.
[0061] The image acquisition module of this embodiment can capture high-definition images of express parcels at different heights passing through the conveyor belt, which is very beneficial for subsequent face sheet extraction and text recognition; the face sheet extraction module uses the deep learning model Unet, which can accurately segment and extract face sheets ; The image correction based on affine transformation can correct the rotation of images from different angles; the binary classification network can accurately classify upright and inverted images, and rotate the inverted image by 180 degrees to obtain an upright image; The location of the receiving address and sorting code adopts the deep learning model YOLO, which c...
Embodiment 3
[0063] Such as image 3 As shown, a logistics bill identification device includes: a conveyor belt 1, a detection bracket 2, a camera sensor, a light source 3, a camera 4 and a controller.
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More