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

Image recognition method and system based on capsule parameter optimization

An image recognition and capsule technology, applied in the field of image recognition, can solve the problems of feature understanding discount, waste of resources, high cost, etc., and achieve the effect of reducing the complexity of time and space, reducing training time, and reducing the amount of calculation

Active Publication Date: 2021-08-03
RENMIN UNIVERSITY OF CHINA
View PDF11 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

CNN discards these attribute features to maximize its feature detection ability, but it is greatly compromised in feature understanding
At the same time, CNN needs a large amount of image data to train the model, and it takes a lot of space to save a copy of all the data, resulting in a lot of waste of resources

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
  • Image recognition method and system based on capsule parameter optimization
  • Image recognition method and system based on capsule parameter optimization
  • Image recognition method and system based on capsule parameter optimization

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0054] Such as figure 1 As shown, the image recognition method based on capsule parameter optimization provided in this embodiment includes the following steps:

[0055] Step S1: Perform filtering operation on the input image data in the primary convolution layer, and perform primary feature extraction on the data. The pixels are converted into local feature outputs through the primary convolutional layer. Preferably, the activation function of this layer uses Relu.

[0056] Specifically, the above-mentioned process of performing primary feature extraction on data includes:

[0057] Step S11: Multiple primary convolutional layers are used to perform filtering operations on the image to obtain primary features of the image. This layer does not use pooling operations, but only convolution operations.

[0058] Step S12: Transform and reorganize the acquired primary features to form primary capsule layers. The primary capsule layer is a convolutional layer with neurons as the ta...

Embodiment 2

[0086]The first embodiment above provides an image recognition method based on capsule parameter optimization, and correspondingly, this embodiment provides an image recognition system. The image recognition system provided in this embodiment can implement the image recognition method based on capsule parameter optimization in Embodiment 1, and the system can be realized by software, hardware or a combination of software and hardware. For example, the system may include integrated or separate functional modules or functional units to execute corresponding steps in the methods of the first embodiment. Since the image recognition system of this embodiment is basically similar to the method embodiment, the description process of this embodiment is relatively simple. For relevant parts, please refer to the part of the description of Embodiment 1. The image recognition system of this embodiment is only schematic .

[0087] The image recognition system based on capsule parameter op...

Embodiment 3

[0093] This embodiment provides a processing device that implements the image recognition method based on capsule parameter optimization provided in Embodiment 1. The processing device may be a processing device for a client, such as a mobile phone, a notebook computer, a tablet computer, a desktop computer, etc. , to execute the image recognition method in Embodiment 1.

[0094] The processing device includes a processor, a memory, a communication interface and a bus, and the processor, the memory and the communication interface are connected through the bus to complete mutual communication. A computer program that can run on the processor is stored in the memory, and the processor executes the image recognition method based on capsule parameter optimization provided in Embodiment 1 when running the computer program.

[0095] Preferably, the memory may be a high-speed random access memory (RAM: Random Access Memory), and may also include a non-volatile memory (non-volatile me...

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 relates to an image recognition method and system based on capsule parameter optimization, and the method comprises the steps: S1, carrying out the filtering operation of an input image through a primary convolution layer, and carrying out the primary feature extraction, and obtaining a primary capsule; S2, designing a capsule block convolution layer, and predicting a high-level capsule by using a low-level capsule through taking a 3D convolution kernel as a conversion matrix to obtain characteristics of complex data; S3, performing whole-group deformation on the features of the complex data to form low-level capsules, screening the low-level capsules by adopting a pruning optimization strategy, and predicting high-level capsules based on the screened low-level capsules; and S4, classifying the images through the digital capsule. According to the method, the identification problem in a complex image can be simply and efficiently processed, the complexity of time and space is reduced, and the method can be widely applied to image identification.

Description

technical field [0001] The invention relates to the technical field of image recognition, in particular to an image recognition method and system based on capsule parameter optimization. Background technique [0002] With the advent of the era of big data and the improvement of hardware computing capabilities, image recognition technology has developed rapidly. But looking at the bigger picture, it's still early days. While adapting to the development of the world, there are also problems such as applicability. Deep learning has achieved excellent results in the field of image recognition, breaking the limitations of traditional pattern recognition and machine learning methods. However, one of the major problems faced by deep networks is that the network structure is becoming more and more complex, which will lead to more and more abstract network structures. Although a complex network can bring strong performance in image recognition, it is also accompanied by some negat...

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): G06K9/62G06N3/04G06N3/08
CPCG06N3/082G06N3/045G06F18/2415Y02T10/40
Inventor 梁循郑香平付虹蛟
Owner RENMIN UNIVERSITY OF CHINA