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

A method and device for computing using volume R-CNN neural network

A neural network and network technology, applied in the field of image processing, can solve the problems of large error in calculation results, occlusion of side view objects, and high requirements for input photos, and achieve the effects of improving prediction accuracy, avoiding occlusion, and powerful computing power.

Active Publication Date: 2021-03-23
INST OF COMPUTING TECH CHINESE ACAD OF SCI
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The problem with the above technology is that the operation is complex and the requirements for input photos are high; the side view is prone to occlusion of objects
Using the focal length to predict the object's length, width, and height parameters may have certain deviations for different mobile phones. Using the formula method to calculate the object volume is not suitable for objects with irregular shapes, and the calculation results have large errors.

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 method and device for computing using volume R-CNN neural network
  • A method and device for computing using volume R-CNN neural network
  • A method and device for computing using volume R-CNN neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0053] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings.

[0054] The invention discloses a calculation method using a Volume R-CNN neural network, extracts and processes key features of an image by using a neural network algorithm, and identifies the types of objects in the image and the volume ratios of various objects.

[0055] The input image includes multiple top-down photos of the same sample from different angles.

[0056] In the processing stage of a single image, the processor uses the improved Faster Region Convolutional Neural Network (faster R-CNN) network to calculate the input image, and marks each object category (class) in the image, and each object border (bounding box), and the predicted volume ratio (volume) for each object. Where volume is expressed as two deci...

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 method and device that uses Volume R‑CNN neural networks, including: multiple images from different angles of the same book; use RPN to determine the recommendation area of sample detection;Border frame; according to the border of the predicted object, the volume ratio occupied by the Volume R‑cnn predicts the volume ratio of each object; according to the type of object category predicted by Fast R‑CNN and the volume ratio of the object volume predicted by Volume R‑CNN, different category objects are calculatedThe volume ratio; among them, the RPN, FAST R‑cnn and Volume R‑cnn shared convolution layer.The present invention can measure complex and various types of samples. By adopting artificial neural network technology and chips, the identification of different types of objects is more accurate and faster.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a method and device for performing calculations using a Volume R-CNN neural network. Background technique [0002] In actual life and production, it is often necessary to measure the volume of various objects in a sample including different types of objects, but there is no quick and accurate measurement method yet. One of the existing technologies is to take the top view and side view of the measurement sample by mobile phone, identify the object type through artificial neural network, and calculate the volume of each object according to the formula. [0003] The problem with the above technology is that the operation is complex and the requirements for input photos are high; the side view is prone to occlusion of objects. Using the focal length to predict the object's length, width, and height parameters may have certain deviations for different mobile phones. Using t...

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): G06N3/04G06N3/08G06K9/62G06F17/15
CPCG06F17/15G06N3/084G06N3/045G06F18/214
Inventor 张团陈云霁
Owner INST OF COMPUTING TECH CHINESE ACAD OF SCI
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