Check patentability & draft patents in minutes with Patsnap Eureka AI!

Bill classification method and system for few samples

A classification method and bill technology, applied in neural learning methods, instruments, biological neural network models, etc., can solve problems such as difficulty in bill classification, and achieve the effect of fast and accurate bill classification

Pending Publication Date: 2020-02-14
西安网算数据科技有限公司
View PDF11 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The present invention proposes a bill classification method and system with a small number of samples, which solves the problem of difficulty in classifying bills with a small number of samples in the prior art

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
  • Bill classification method and system for few samples
  • Bill classification method and system for few samples
  • Bill classification method and system for few samples

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0046] Such as figure 1 As shown, a small-sample bill classification method includes

[0047] S0: Construct bill sample set A, the bill types in bill sample set A are n, n≥30, the number of each bill type is m, m≥15,

[0048] A={i_j.jpg|i=1, 2...n; j=1, 2...m};

[0049] S1: Compare the bill pictures in the bill sample set A in pairs to form a paired bill sample set S:

[0050] S={(k1.jpg, k2.jpg, y k )|k1.jpg∈A, k2.jpg∈A}

[0051] Among them, y k is a similar label, when k1.jpg and k2.jpg are of the same ticket type, y k = 0, otherwise y k = 1;

[0052] S2: According to the paired bill sample set S, construct a bill picture similarity matching model C, including

[0053] S21: Build a two-way deep learning network, including the first deep learning network and the second deep learning network. The first deep learning network obtains the binary hash code b of k1.jpg k1 , The second deep learning network obtains the binary hash code b of k2.jpg k2 ,

[0054] S22: Accor...

Embodiment 2

[0074] Based on the same inventive concept as in Embodiment 1, this embodiment proposes a bill classification system with a small number of samples, including

[0075] The first obtaining unit 21 is configured to obtain a bill picture and construct a bill sample set A;

[0076] The first calculation unit 22 is used to construct a two-way deep learning network, and calculate the binary hash code of the bill picture;

[0077] The second calculation unit 23 is used to calculate the loss function of the two-way deep learning network;

[0078] The first judging unit 24 is used to judge whether the training of the bill picture similarity model is completed;

[0079] The third calculation unit 25 is used to calculate the binary hash code of the newly added note picture;

[0080] The fourth calculation unit 26 is used to calculate the distance between the newly added binary hash code of the bill picture and the binary hash code of each bill picture in the bill picture similarity mat...

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 belongs to the technical field of intelligent account making, and provides a bill classification method and system for a small number of samples. The method comprises the following steps: obtaining bill pictures, and constructing a bill picture set A; constructing a paired bill sample set S; constructing a bill picture similarity matching model C; and loading the trained bill picturesimilarity matching model C, and judging the bill category of the newly added bill picture. By means of the technical scheme, the problem that in the prior art, bills of a small number of samples aredifficult to classify is solved.

Description

technical field [0001] The invention belongs to the technical field of intelligent accounting, and relates to a bill classification method and system for a small number of samples. Background technique [0002] During the actual operation of the bill system, it will accept bill pictures from various companies and different industries. The format of these bill pictures is different and there are many types. Before the system is processed, it is often necessary to classify according to the bill style, that is to say, bill samples of the same type or format are classified into one category, and bills of different types are classified into different categories. If traditional computer vision or machine learning technology requires A large number of bill pictures are used as training samples to build a machine learning model, and the large number mentioned here is often hundreds to thousands of pictures of each type of bill sample, which is generally difficult to meet in real sce...

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
CPCG06N3/08G06N3/045G06F18/22G06F18/241G06F18/214
Inventor 张汉宁苏斌弋渤海徐博田福康
Owner 西安网算数据科技有限公司
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com