Unlock instant, AI-driven research and patent intelligence for your innovation.

Grain quality detection method based on transfer learning and adaptive deep convolutional neural network

A deep convolution and transfer learning technology, applied in neural learning methods, biological neural network models, neural architectures, etc., can solve problems such as time-consuming and expensive, and unrealistic models

Active Publication Date: 2020-06-05
JIANGNAN UNIV
View PDF8 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, collecting a large number of labeled samples in a new environment is time-consuming and expensive, and it is unrealistic to train a new model again

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
  • Grain quality detection method based on transfer learning and adaptive deep convolutional neural network
  • Grain quality detection method based on transfer learning and adaptive deep convolutional neural network
  • Grain quality detection method based on transfer learning and adaptive deep convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0050] The present invention will be further described below in conjunction with specific drawings and embodiments.

[0051] This application discloses a grain quality detection method based on migration learning and adaptive deep convolutional neural network. The method includes the following steps, please refer to figure 1 The flowchart shown:

[0052] Step S1: Build an image acquisition system to collect samples in different fields, select a uniformly illuminated black background as the source field, and collect M source field samples {X S ,Y S}; select the white background with uneven illumination as the target area, and collect N target area samples {X T ,Y T}, the samples in both domains include qualified samples and defective samples, both M and N are positive integers, and M>N. Segment all sample images and unify their sizes. The specific method can refer to the existing method, which will not be repeated in this application. The source domain samples and target d...

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 grain quality detection method based on transfer learning and an adaptive deep convolutional neural network, and relates to the field of machine vision and grain quality detection. The method comprises the following steps: acquiring grain sample images in the source field and the target field, wherein the images comprises a sample image of qualified grains and a sample image of grains with defects; and selecting a deep convolutional neural network CNN model to identify defects, initializing the CNN model by using model parameters trained in the source domain, and introducing a transfer learning algorithm to assist a target domain sample by using a source domain sample to complete quality detection of grains in the target domain. An adaptive learning rate is provided in training of the CNN model, a quadratic function and a normal distribution model are introduced, model parameters are updated in a gradient descent mode and a gradient ascent mode respectively, and model loss is optimized. According to the method, the training performance of the CNN model can be improved, the change of the field is self-adapted and the accuracy of grain quality detection is greatly improved.

Description

technical field [0001] The invention relates to the field of machine vision and grain quality detection, in particular to a grain quality detection method based on migration learning and adaptive deep convolutional neural network. Background technique [0002] Computer vision technology provides a real-time, efficient, and non-destructive detection method, which is usually combined with intelligent algorithms to obtain representative features of the measured object. Convolutional neural network (CNN) is a deep recognition model that provides an "end-to-end" learning method. The input original image is mapped to a feature representation set through the feature layer, and then multi-classification is realized through the classification layer at the end. Therefore, the CNN quality detection method based on computer vision technology is a fully automatic intelligent detection method that does not depend on prior knowledge. [0003] There are still many problems in CNN in practi...

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 Applications(China)
IPC IPC(8): G01N21/88G06N3/04G06N3/08
CPCG01N21/8851G06N3/084G01N2021/8883G01N2021/8887G06N3/045
Inventor 李可张思雨张秋菊
Owner JIANGNAN UNIV