sequential image prediction method based on LSTM and DCGAN

A prediction method and time series prediction technology, applied in neural learning methods, character and pattern recognition, instruments, etc., to achieve accurate prediction, solve high-dimensional incomputability, and reduce feature dimensions.

Active Publication Date: 2019-05-31
NANJING UNIV OF INFORMATION SCI & TECH
View PDF5 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But the traditional RNN has long-term dependencies (Long-Term Dependencies)

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
  • sequential image prediction method based on LSTM and DCGAN
  • sequential image prediction method based on LSTM and DCGAN
  • sequential image prediction method based on LSTM and DCGAN

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0029] The technical solutions of the present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0030] The temporal image prediction method based on LSTM and DCGAN of the present invention comprises steps:

[0031] (1) Construct a DCGAN encoder, including an encoding module and a decoding module, and an LSTM network for learning temporal images is connected between the two modules to predict the feature distribution;

[0032] In the encoding module, a network structure of four-layer convolution and four-layer downsampling is designed; in the decoding module, four-layer deconvolution and four-layer upsampling are used; the LSTM network for learning time series images is connected between the two modules to predict features distributed. Such as figure 1 As shown, first collect images and input them into the encoding module to extract spatial features; input the extracted features into LSTM for prediction, and pass the pr...

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 discloses a sequential image prediction method based on LSTM and DCGAN, and the method combines the excellent feature capture capability of the DCGAN with the LSTM, can enable the predicted image data to be visualized, and facilitates the direct observation. The improved LSTM network has convolution characteristics inside, and two-dimensional spatial characteristics of image data canbe directly learned; In order to reduce the internal learning complexity, a traditional input image is changed into an input feature; Characteristics are extracted from DCGAN, and compared with an original image, the method has the advantages that the dimension is greatly simplified, and the whole network is controllable. According to the method, the feature dimension is well reduced through theDCGAN, and the problem that the high dimension cannot be calculated is solved; The improved LSTM can better learn time sequence characteristics, so that more accurate prediction is realized; The wholenetwork structure complies with a stack type cascading strategy in the connection method, and guarantees are provided for controlling the network depth. The sequential image prediction model architecture provided by the invention is theoretically suitable for all sequential images.

Description

technical field [0001] The invention relates to an improved method for temporal image prediction, in particular to a temporal image prediction method based on LSTM and DCGAN. Background technique [0002] At this stage, the combination of image recognition and deep learning has become a research hotspot in computer vision, but as far as the current development situation is concerned, there are still great limitations. The greatest extent is that the recognized objects are discrete and unrelated to each other. , and mainly based on classification. In order to better expand related business needs, the recent development of image recognition has focused on the timing of the images associated with each other. Through effective learning, the change of image features at a specified time in the future can be predicted, and the traditional classification operation can be extended to the prediction operation. The study of temporal images can benefit in a variety of applications, su...

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
Inventor 方巍张飞鸿丁叶文
Owner NANJING UNIV OF INFORMATION SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap