Bird identification method based on deep learning

A technology of deep learning and recognition methods, applied in the field of bird recognition based on deep learning algorithms, can solve the problems of slow speed and low accuracy of bird images, and achieve faster recognition speed, good robustness, and low time complexity Effect

Inactive Publication Date: 2018-09-07
PEKING UNIV SHENZHEN GRADUATE SCHOOL
View PDF4 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a bird recognition method based on deep learning. By establishing a two-stage training framework, a classification model for specific birds in mangrove coast...

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
  • Bird identification method based on deep learning
  • Bird identification method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0016] Below in conjunction with accompanying drawing, further describe the present invention through embodiment, but do not limit the scope of the present invention in any way.

[0017] The present invention provides a bird recognition method based on deep learning. For different data sets and new data sets, the original network model can be expanded through fine-tuning, which can speed up the bird recognition speed of a single image and improve the accuracy of bird recognition. Rate.

[0018] figure 1 It is a block flow diagram of a bird recognition method based on deep learning provided by an embodiment of the present invention, including the following steps:

[0019] A. Collect a dataset of images of mangroves and coastal wetlands, flip them horizontally to double the amount of data, and use it as a training dataset;

[0020] B. Based on the ImageNet pre-trained deep network, the neural network can classify 1000 categories of imagenet; modify the ImageNet pre-trained dee...

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 bird identification method based on deep learning. Two stages of training frame are establishes. In the first stage, a deep network trained on a large scale data set is usedfor training a CUB-200-2011 bird dataset to obtain a fine tuning model for the bird dataset; and in the second stage, the model of the first stage is subjected to fine tuning on a self-built mangroveforest coastal wetland dataset, a classification model for the specific birds of the mangrove forest coastal wetland is obtained, and the classification model is used for realizing bird identificationbased on deep learning. By use of the method, the bird identification based on the deep learning can be realized, and the problems of low speed and excessively low accuracy when bird images are classified and identified by an existing machine learning method are solved.

Description

technical field [0001] The invention belongs to the technical field of machine learning and bird recognition, relates to a bird recognition technology based on image classification, and in particular to a bird recognition method based on a deep learning algorithm. Background technique [0002] In recent years, with the development of deep learning, the accuracy of image classification using deep learning algorithms on specific data sets (such as ImageNet, CUB-200-2011) has been continuously improved, and image classification has risen from the field of scientific research to specific practical applications. , specifically including image recognition, image retrieval (searching images by image) and so on. In the field of image classification, images in natural scenes are the main source of the database, and bird recognition is one of the very important application scenarios. [0003] The essence of bird recognition is the classification of bird images. At present, the exist...

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): G06K9/00G06K9/62G06N3/04
CPCG06V20/38G06N3/045G06F18/214
Inventor 邹月娴陆超豪
Owner PEKING UNIV SHENZHEN GRADUATE SCHOOL
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