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

A method for authenticity identification of cultural relics based on knowledge distillation

A technology for authenticity identification and cultural relics, which is applied in the field of cultural relics authenticity identification based on knowledge distillation, which can solve the problems of time-consuming manual identification process and a large investment of manpower and material resources, so as to save identification costs, improve detection efficiency, and improve accuracy Effect

Active Publication Date: 2021-03-26
SHENZHEN GRADUATE SCHOOL TSINGHUA UNIV
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Such a manual identification process takes a long time, requires a lot of manpower and material resources, and the identification results are highly subjective.

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
  • A method for authenticity identification of cultural relics based on knowledge distillation
  • A method for authenticity identification of cultural relics based on knowledge distillation
  • A method for authenticity identification of cultural relics based on knowledge distillation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0048] In this example, 5 fingerprint areas are known before the cultural relics are exhibited, and the size is 5×5mm 2 . Under different lighting conditions, use EOS 7D Mark II camera equipped with MP-E 65mm f / 2.8 1-5X lens to collect 500 RGB images from multiple angles in 5 fingerprint areas, with a resolution of 5472×3648 pixels. A total of 2500 images were obtained. Use the labeling tool labelimg to label the fingerprint area in the image, and get the XML file corresponding to each image. Randomly select 2300 images and their annotation files as the training set, and the rest as the verification set.

[0049] Step 2: Configure the YOLOV3 network as the teacher network, configure the YOLOV3-Tiny network as the student network; modify the network model and parameters;

[0050] In this embodiment, the deep learning development environment needs to be configured, the CPU is i79700K, the GPU is NVIDIAGeForceRTX2080, the operating system is Ubuntu 16.04LTS, CUDA10.0, and the ...

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 present invention relates to the field of authenticity identification of cultural relics, and in particular to a method for authenticity identification of cultural relics based on knowledge distillation, which mainly includes: step 1: before exhibiting cultural relics, collect fingerprint area images and make data sets; step 2: configure YOLOV3 network as For the teacher network, configure the YOLOV3-Tiny network as the student network; Step 3: Train YOLOV3; Step 4: Train YOLOV3-Tiny based on knowledge distillation; Step 5: After the cultural relics are recovered, re-collect the fingerprint area image to make a test set; Step 6: Use The trained YOLOV3‑Tiny identifies the authenticity of cultural relics. The present invention uses YOLOV3 with good accuracy but slow speed as the teacher network, and YOLOV3-Tiny with poor accuracy but fast speed as the student network to carry out knowledge distillation, supervise the students' network learning with the softened target, and maintain the original YOLOV3-Tiny With a faster detection speed, its accuracy is greatly improved, the hardware resource occupation in the cultural relics identification process is reduced, the detection efficiency is improved, and the identification cost is saved.

Description

technical field [0001] The invention relates to the field of authenticity identification of cultural relics, in particular to a method for authenticity identification of cultural relics based on knowledge distillation. Background technique [0002] Our country has a long history and culture, and there are a large number of cultural relics. In order to promote history and culture, cultural relics collection units all over the country will regularly carry out cultural relics tour activities. After the event is exhibited, the cultural relics need to be identified to prevent the cultural relics from being destroyed or replaced. At present, the identification of cultural relics is mainly done manually, which depends on the personal experience and knowledge range of experts, and sometimes requires high-tech equipment to assist in detection. Such a manual identification process takes a long time, requires a lot of manpower and material resources, and the identification results ar...

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 Patents(China)
IPC IPC(8): G06N3/08G06Q50/10G06K9/62
CPCG06N3/08G06Q50/10G06F18/23213
Inventor 刘学平李玙乾张晶阳王哲
Owner SHENZHEN GRADUATE SCHOOL TSINGHUA UNIV