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

Multi-angle video fusion commodity identification algorithm for unmanned vending counter

A video fusion and recognition algorithm technology, applied in character and pattern recognition, computing, computer parts and other directions, can solve the problems of limited information, low recognition accuracy, easy to be blocked, etc., to enrich effective information and reduce redundancy The amount of data, the effect of improving the efficiency and accuracy of product identification

Active Publication Date: 2018-12-07
武汉市哈哈便利科技有限公司
View PDF7 Cites 36 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in this patent, the video stream data only comes from a single angle, so the information contained is limited, and it is especially vulnerable to occlusion, resulting in low recognition accuracy

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
  • Multi-angle video fusion commodity identification algorithm for unmanned vending counter

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0036] Such as figure 1 As shown, this embodiment provides a product recognition algorithm for multi-angle video fusion of unmanned vending cabinets. The algorithm model is mainly modeled based on the Keras / TensorFlow deep learning framework. The model is trained, and once the model training is completed, the model can be applied to predict new input video data to obtain a final recognition result.

[0037] The specific implementation steps are as follows:

[0038] Step 1. Collect video data: Three cameras at different angles (view angle 1, view angle 2, and view angle 3) on the container record the user’s single purchase operation, and the recorded video stream data (video stream 1, video stream 2. Video stream 3) respectively number video1, video2 and video3 and place them in the same folder, and at the same time, this folder contains information about the type of goods purchased by the user in this operation and the corresponding quantity of goods, formatted as .json The ...

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

The invention provides a multi-angle video fusion commodity identification algorithm for an unmanned vending counter. The algorithm comprises the following steps: pre-processing collected video data;performing key frame sampling on video frames in the video data; building a deep convolutional neural network model; and performing training and precision testing on the model by utilizing the sampledkey frames, and, after accuracy of commodity identification reaches expectation, deploying the model. The algorithm further comprises: feature extraction, feature fusion, feature identification. Thealgorithm further comprises the step that the video data from cameras at different angles of the unmanned vending counter are subjected to pre-processing and key frame sampling and then are input intothe model to obtain identified commodity types and identified corresponding commodity quantities. Through the multi-angle video fusion technology, rich information brought by multi-source data is fully used to reduce the impact of blocking goods, and then the accuracy of commodity identification is improved.

Description

technical field [0001] The invention relates to the technical field of unmanned vending cabinets, in particular to a commodity recognition algorithm for multi-angle video fusion of unmanned vending cabinets. Background technique [0002] With the development of artificial intelligence technology, all walks of life have begun to apply artificial intelligence technology to reduce industry operating costs and improve their efficiency, especially in the field of new retail, how to use artificial intelligence technology to reduce operating costs and make goods within reach, convenient It has become a hot research field in the industry. With the breakthroughs made by researchers in the field of computer vision in recent years, it has become completely feasible to use image recognition technology based on deep learning neural networks to automatically identify products purchased by customers. On the other hand, due to the improvement of computer computing power, this technology is...

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
IPC IPC(8): G06K9/00G06K9/62G06N3/08
CPCG06N3/08G06V20/41G06V20/46G06F18/253
Inventor 蔡丁丁方无迪唐开刘钰涛张运辉
Owner 武汉市哈哈便利科技有限公司
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