Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A logistics composite code recognition method based on multi-task deep learning

A technology of deep learning and identification methods, which is applied in the fields of collaborative operation devices, biological neural network models, record carriers used by machines, etc., and can solve problems such as positioning and identification of logistics composite codes that are not involved.

Active Publication Date: 2021-04-06
ZHEJIANG HANQIANG AUTOMATION EQUIP
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At the same time, it does not involve the positioning and identification of logistics composite codes

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A logistics composite code recognition method based on multi-task deep learning
  • A logistics composite code recognition method based on multi-task deep learning
  • A logistics composite code recognition method based on multi-task deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment approach 1

[0088] refer to Figure 1 to Figure 10 , a logistics composite code recognition system based on multi-task deep learning, such as Figure 10 As shown, the main process is as follows: when the goods to be sorted on the assembly line move to a certain station of the assembly line, the system automatically triggers multiple cameras to take images of the goods from various directions; then, based on the Faster R-CNN goods composite code label detection Positioning to obtain the composite code label image on the cargo image; then, perform affine transformation on the minimum area rectangle of the composite label image; further, segment the image of the composite label after the affine transformation, and segment the label image of the inkjet characters And a one-dimensional barcode image or a two-dimensional barcode image; then, the label image of the obtained inkjet character is used for character recognition based on Faster R-CNN, and the character sequence in the label image of t...

Embodiment approach 2

[0137] The above is similar to Embodiment 1, except that there is no special requirement for the placement of goods on the assembly line.

[0138] Since the face of the composite code label pasted on the goods is not required to face up in this embodiment, it is necessary to adopt a method for obtaining the image of the composite code label in all directions, such as Figure 9 Shown; For express logistics, the goods are on the sorting line, the size of the goods and the position angle of the composite code label pasted on the goods are all uncertain; in order to obtain the image of the composite code label from various angles, in the present invention Five cameras are installed on the top of the assembly line, left, right, front and back, so that images of the composite code labels in all directions of the goods can be captured; in addition, in order to ensure that the composite code labels pasted on the goods can be read accurately, the present invention stipulates that Paste...

Embodiment approach 3

[0140] The above is similar to Embodiment 1, the difference is the positioning and division of the composite code label. In this embodiment, the Faster R-CNN network is used to locate and segment the label part of the coded characters in the composite code, and then according to the position information of the label part of the coded characters and the label part of the coded characters and the one-dimensional barcode or The positional relationship information of the label portion of the two-dimensional barcode is further positioned to segment the label portion of the one-dimensional barcode or the two-dimensional barcode in the composite code label.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

A logistics composite code recognition method based on multi-task deep learning, including a method for comprehensively obtaining logistics composite code images, a label design scheme for visual detection and positioning of sprayed code characters, character positioning and Segmented character size, glyph and space between characters design scheme, composite code design scheme suitable for visual recognition, Faster R-CNN network for detecting and positioning composite codes, composite code image deviation correction and character Algorithm module for upside-down detection, multi-task deep convolutional neural network for deep learning and training recognition, convolutional neural network for character recognition on labels of inkjet characters based on deep learning, and one-dimensional recognition for composite codes The barcode algorithm module, the algorithm module used to identify the two-dimensional barcode in the composite code, and the sorting control module used to control the sorting action according to the recognized composite code information. The invention effectively solves the problem that a large number of randomly placed, irregularly shaped and flexibly packaged goods cannot be quickly and automatically sorted.

Description

technical field [0001] The invention relates to the application of artificial intelligence, digital image processing, convolutional neural network and computer vision in the identification and sorting of logistics composite code labels, and belongs to the field of intelligent logistics. Background technique [0002] In the current logistics system, the code reading equipment of the system is mainly for any kind of information identification in spray code characters, one-dimensional code, two-dimensional code, and RFID. The way to obtain product and logistics carrier information is to obtain them separately, and then manually assemble the product information on the computer or PAD and bind it with the information of the logistics carrier, so as to ensure the effective management of the goods in the process of transportation and storage. [0003] Because the cost is very low, inkjet characters, one-dimensional barcodes and two-dimensional barcodes have been widely used in the ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06K19/06G06K17/00G06N3/04B07C3/18
CPCG06K17/0022G06K19/06028G06K19/06037B07C3/18G06N3/045
Inventor 盛力峰关亮林宏鋆盛雷雷王路贾宝荣王显杰聂学雯王权
Owner ZHEJIANG HANQIANG AUTOMATION EQUIP
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com