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

Image ancient poetry generation method based on Faster R-convolutional neural network detection model

A convolutional neural network and detection model technology, applied in the computer field, can solve problems such as the lack of image data sets of ancient poems, the limitation of text expression ability, and the lack of judgment of the emotional tendency of ancient poems, so as to achieve rich functionality and readability, Improve the quality and fun of generation, and promote the effect of traditional culture

Inactive Publication Date: 2022-06-24
SHAANXI NORMAL UNIV
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the mainstream text input method based on deep learning has big problems: First, it is limited to the text expression ability of the input text, and the input of a small number of keywords makes the generated ancient poems unable to fully express the user's writing intention and emotional fluctuations.
Second, the existing image data sets have the problems of low detection accuracy and slow detection speed when faced with the generation of image ancient poems, and there is a lack of image data sets dedicated to ancient poetry image words
Third, there is a lack of judgment on the emotional orientation of ancient poems

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 ancient poetry generation method based on Faster R-convolutional neural network detection model
  • Image ancient poetry generation method based on Faster R-convolutional neural network detection model
  • Image ancient poetry generation method based on Faster R-convolutional neural network detection model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0061] exist figure 1 , the image ancient poem generation method based on the Faster R-convolutional neural network detection model of the present embodiment consists of the following steps:

[0062] (1) Collect pictures of ancient poetic imagery words

[0063] Based on 100 common image words in ancient poems, the crawler method was used to crawl 100 pictures corresponding to the image words from the Internet image data, and a total of 10,000 images of ancient poems were obtained.

[0064] (2) Image preprocessing of ancient poetry and imagery

[0065] The size of the collected image word pictures is unified, and the piecewise linear grayscale enhancement method is used to process the details grayscale of the pictures to enhance the image contrast and compress unnecessary image details.

[0066] The detail gray level processing of the picture by the piecewise linear gray level enhancement method is as follows: the gray level of the image f(x, y) input by the user is from 0 to...

Embodiment 2

[0118] The image ancient poem generation method based on the Faster R-convolutional neural network detection model of the present embodiment consists of the following steps:

[0119] (1) Collect pictures of ancient poetic imagery words

[0120] This procedure is the same as in Example 1.

[0121] (2) Image preprocessing of ancient poetry and imagery

[0122] The size of the collected image word pictures is unified, and the piecewise linear grayscale enhancement method is used to process the details grayscale of the pictures to enhance the image contrast and compress unnecessary image details.

[0123] The detail gray level processing of the picture by the piecewise linear gray level enhancement method is as follows: the gray level of the image f(x, y) input by the user is from 0 to 32 levels, and the image f(x, y) in this embodiment is ) is level 0, the specific level of the gray level of the image f(x, y) should be determined according to the input image, and the gray level...

Embodiment 3

[0143] The image ancient poem generation method based on the Faster R-convolutional neural network detection model of the present embodiment consists of the following steps:

[0144] (1) Collect pictures of ancient poetic imagery words

[0145] This procedure is the same as in Example 1.

[0146] (2) Image preprocessing of ancient poetry and imagery

[0147] The size of the collected image word pictures is unified, and the piecewise linear grayscale enhancement method is used to process the details grayscale of the pictures to enhance the image contrast and compress unnecessary image details.

[0148] The detail gray level processing of the picture by the piecewise linear gray level enhancement method is as follows: the gray level of the image f(x, y) input by the user is from 0 to 32 levels, and the image f(x, y) in this embodiment is ) is 32, and the specific gray level of the image f(x, y) should be determined according to the input image, and the gray level of the enhanc...

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 an image ancient poetry generation method based on a Faster R-convolutional neural network detection model. The method comprises the steps of collecting ancient poetry image word pictures, preprocessing the ancient poetry image word pictures, constructing an ancient poetry image word image data set, inputting user images, extracting image keyword features, extracting visual image features, constructing an ancient poetry text generation model, judging the emotional tendency of the ancient poetry, and displaying and generating the ancient poetry. According to the method, the ancient poetry image word image data set is constructed through collection and training, so that the accuracy, the detection speed and the generation speed of image detection during image ancient poetry generation are improved. The image keyword features and the image visual features are combined in the ancient poetry generation network, and the theme consistency of the picture and the ancient poetry is improved. The emotional tendency of the generated ancient poetry is judged in the generation of the image ancient poetry, the generation function of the image ancient poetry is enriched, and the generation quality of the image ancient poetry is improved. The method has the advantages of high generation speed, high consistency of images and ancient poetry themes and the like, and can be used in the technical field of image ancient poetry generation.

Description

technical field [0001] The invention belongs to the field of computer technology, and specifically relates to computer image target detection, natural language generation and text emotion classification. Background technique [0002] Ancient poetry generation is an important and challenging research task in the field of natural language generation, aiming to enable computers to create high-quality poems like poets. The automatic generation of poetry has undergone a transition from traditional machine translation methods to deep learning text input methods. However, the mainstream text input methods based on deep learning have major problems: First, limited to the textual expression ability of the input text, the input of a small number of keywords makes the generated ancient poems unable to fully express the user's writing intentions and emotional fluctuations. Second, the existing image data sets have the problems of low detection accuracy and slow detection speed when fac...

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): G06F40/166G06F40/30G06V10/40G06K9/62G06N3/04G06N3/08G06F16/951G06V10/774G06V10/82
CPCG06F40/166G06F40/30G06F16/951G06N3/08G06N3/047G06N3/045G06F18/214
Inventor 谈启雷吴晓军杨红红张玉梅
Owner SHAANXI NORMAL UNIV