scholarly journals RoBINN: Robust Bird Species Identification using Neural Network

2021 ◽  
Author(s):  
Chirag Samal ◽  
Prince Yadav ◽  
Sakshi Singh ◽  
Satyanarayana Vollala ◽  
Amrita Mishra
Keyword(s):  
Neural Network ◽  
Species Identification ◽  
Bird Species

RoBINN: Robust Bird Species Identification using Neural Network

2021 ◽  
Author(s):  
Chirag Samal ◽  
Prince Yadav ◽  
Sakshi Singh ◽  
Satyanarayana Vollala ◽  
Amrita Mishra
Keyword(s):  
Neural Network ◽  
Species Identification ◽  
Bird Species

scholarly journals Bird Species Identification Using Spectrogram Based on Multi-Channel Fusion of DCNNs

Entropy ◽  
2021 ◽  
Vol 23 (11) ◽  
pp. 1507
Author(s):  
Feiyu Zhang ◽  
Luyang Zhang ◽  
Hongxiang Chen ◽  
Jiangjian Xie
Keyword(s):  
Species Identification ◽  
Feature Fusion ◽  
Bird Species ◽  
The Other ◽  
Identification Accuracy ◽  
Mean Average Precision ◽  
Training Dataset ◽  
Model Parameters ◽  
Average Precision ◽  
Fusion Mode

Deep convolutional neural networks (DCNNs) have achieved breakthrough performance on bird species identification using a spectrogram of bird vocalization. Aiming at the imbalance of the bird vocalization dataset, a single feature identification model (SFIM) with residual blocks and modified, weighted, cross-entropy function was proposed. To further improve the identification accuracy, two multi-channel fusion methods were built with three SFIMs. One of these fused the outputs of the feature extraction parts of three SFIMs (feature fusion mode), the other fused the outputs of the classifiers of three SFIMs (result fusion mode). The SFIMs were trained with three different kinds of spectrograms, which were calculated through short-time Fourier transform, mel-frequency cepstrum transform and chirplet transform, respectively. To overcome the shortage of the huge number of trainable model parameters, transfer learning was used in the multi-channel models. Using our own vocalization dataset as a sample set, it is found that the result fusion mode model outperforms the other proposed models, the best mean average precision (MAP) reaches 0.914. Choosing three durations of spectrograms, 100 ms, 300 ms and 500 ms for comparison, the results reveal that the 300 ms duration is the best for our own dataset. The duration is suggested to be determined based on the duration distribution of bird syllables. As for the performance with the training dataset of BirdCLEF2019, the highest classification mean average precision (cmAP) reached 0.135, which means the proposed model has certain generalization ability.

scholarly journals Local compressed convex spectral embedding for bird species identification

2018 ◽  
Vol 143 (6) ◽  
pp. 3819-3828 ◽  
Author(s):  
Anshul Thakur ◽  
Vinayak Abrol ◽  
Pulkit Sharma ◽  
Padmanabhan Rajan
Keyword(s):  
Species Identification ◽  
Bird Species ◽  
Spectral Embedding

Simple Neural Network Reveals Unexpected Patterns of Bird Species Richness in Forest Fragments

2005 ◽  
Vol 20 (5) ◽  
pp. 513-527 ◽  
Author(s):  
Claude Monteil ◽  
Marc Deconchat ◽  
Gérard Balent
Keyword(s):  
Neural Network ◽  
Species Richness ◽  
Bird Species ◽  
Forest Fragments ◽  
Bird Species Richness ◽  
Simple Neural Network

Plant Identification Based on Artificial Intelligence

2011 ◽  
Vol 255-260 ◽  
pp. 2286-2290 ◽  
Author(s):  
Yang Ying Gan ◽  
Chun Sheng Hou ◽  
Ting Zhou ◽  
Shu Fa Xu
Keyword(s):  
Neural Network ◽  
Artificial Intelligence ◽  
Artificial Neural Network ◽  
Expert System ◽  
Machine Vision ◽  
Species Identification ◽  
Plant Identification ◽  
Artificial Intelligent ◽  
Botanical Research ◽  
Identification Tool

Species identification plays an important role in botanical research, but traditional identification tool, which mainly depends on reference books or identification keys, is often recognized as a difficult and frustrating task, especially for novices. In recent decades, many efforts have been made by taxonomists and programmers to ease the difficulty of species identification by developing a range of tools that increasingly involved the use of computers. In this paper, new advances of plant identification based on three main artificial intelligent technologies: expert system, artificial neural network, and machine vision are briefly introduced. Several trends of plant identification tools for non-expert users are also proposed in the last part.

Sign in / Sign up

Export Citation Format

Share Document