An ultra-specific image dataset for automated insect identification

Abeywardhana, D. L.; Dangalle, C. D.; Nugaliyadde, Anupiya; Mallawarachchi, Yashas

doi:10.1007/s11042-021-11693-3

Cited by 9 publications

(6 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A couple of months ago, Abeywardhana et al [ 43 ] produced a dataset with an unbalanced and restricted number of photos for six taxa of the Cicindelinae subfamily (tiger beetles) of the order Coleoptera. The authors put much effort into photographing tiger beetles from various sources, perspectives, and scales, although they recognize that tiger beetle categorization can be difficult even for a trained human eye.…”

Section: Related Workmentioning

confidence: 99%

Intelligent Identification of Jute Pests Based on Transfer Learning and Deep Convolutional Neural Networks

Sourav

Wang

2022

Neural Process Lett

View full text Add to dashboard Cite

Pest attacks pose a substantial threat to jute production and other significant crop plants. Jute farmers in Bangladesh generally distinguish between different pests that appear to be the same using their eyes and expertise, which isn't always accurate. We developed an intelligent model for jute pests identification based on transfer learning (TL) and deep convolutional neural networks (DCNN) to solve this practical problem. The proposed DCNN model can realize fast and accurate automatic identification of jute pests based on photographs. Specifically, the VGG19 CNN model was trained by TL on the ImageNet database. A well-structured image dataset of four dominant jute pests is also established. Our model shows a final accuracy of 95.86% on the four most vital jute pest classes. The model’s performance is further demonstrated by the precision, recall, F1-score, and confusion matrix results. The proposed model is integrated into Android and IOS applications for practical uses.

show abstract

Section: Related Workmentioning

confidence: 99%

Intelligent Identification of Jute Pests Based on Transfer Learning and Deep Convolutional Neural Networks

Sourav

Wang

2022

Neural Process Lett

View full text Add to dashboard Cite

show abstract

“…In the above formula, the SVM loss function has strict requirements on the network model. In the process of animation image recognition stability analysis, the model can correctly classify and recognize animation images [10]. And when the confidence is high enough, the value of SVM loss will be equal to zero [11].…”

Section: Neural Network Loss Functionmentioning

confidence: 99%

Stability of Neural Networks Dependent on Time Series in Anime Image Recognition

2021

View full text Add to dashboard Cite

Time series and neural networks have profoundly affected the fields of artificial intelligence and machine learning. It can be applied to tasks such as image recognition and target detection with high accuracy. In order to solve the shortcomings of the existing time series neural network animation image recognition stability research, this paper discusses the time series model stationarity, animation image image feature extraction and the functional equation of neural network loss. The dataset and parameter settings for sequential neural network animation image recognition applications are briefly introduced. And the work flow design of the time series neural network animation image recognition system structure model is discussed, and finally the stability of the time series neural network in animation image recognition is compared with the machine learning SVF, FCN model and RFC model. The experimental data show that the average absolute error of the time series neural network model in animation image recognition is small. When the number of iterations is 50-250, the average absolute error of the time series neural network model is less than 2%, while the machine learning SVF, FCN The average absolute error of the model and the RFC model is greater than 10%, so it is verified that the time series neural network model has a faster convergence speed and higher stability in animation image recognition.

show abstract

“…They are of choice and have been extensively used for insect identification involving whole insect image recognition. They demonstrate astonishing accuracy on a wide range of Arthropods 6 – 8 , including Culicidae 9 – 11 . Features related to animal behavior (e.g., flying and walking trajectories, postures, etc.)…”

Section: Introductionmentioning

confidence: 99%

“…Models based on insect morphology imaging of immobilized insects, an approach close to the entomological expertise deployed to identify insects that have inter-genus inter-species high morphological similarities, require a considerable number of data for training each Genus/species to learn the features and gain validation accuracy 10 , 13 , 14 . Databases needed to train such models on whole insect recognition are filled with pictures of several poses, dorsal–ventral, etc., to collect taxonomic discrimant characters 6 , 15 – 17 .…”

Section: Introductionmentioning

confidence: 99%

Wing Interferential Patterns (WIPs) and machine learning for the classification of some Aedes species of medical interest

Cannet,

Simon-Chane,

Histace

et al. 2023

Sci Rep

View full text Add to dashboard Cite

Hematophagous insects belonging to the Aedes genus are proven vectors of viral and filarial pathogens of medical interest. Aedes albopictus is an increasingly important vector because of its rapid worldwide expansion. In the context of global climate change and the emergence of zoonotic infectious diseases, identification tools with field application are required to strengthen efforts in the entomological survey of arthropods with medical interest. Large scales and proactive entomological surveys of Aedes mosquitoes need skilled technicians and/or costly technical equipment, further puzzled by the vast amount of named species. In this study, we developed an automatic classification system of Aedes species by taking advantage of the species-specific marker displayed by Wing Interferential Patterns. A database holding 494 photomicrographs of 24 Aedes spp. from which those documented with more than ten pictures have undergone a deep learning methodology to train a convolutional neural network and test its accuracy to classify samples at the genus, subgenus, and species taxonomic levels. We recorded an accuracy of 95% at the genus level and > 85% for two (Ochlerotatus and Stegomyia) out of three subgenera tested. Lastly, eight were accurately classified among the 10 Aedes sp. that have undergone a training process with an overall accuracy of > 70%. Altogether, these results demonstrate the potential of this methodology for Aedes species identification and will represent a tool for the future implementation of large-scale entomological surveys.

show abstract

An ultra-specific image dataset for automated insect identification

Cited by 9 publications

References 43 publications

Intelligent Identification of Jute Pests Based on Transfer Learning and Deep Convolutional Neural Networks

Intelligent Identification of Jute Pests Based on Transfer Learning and Deep Convolutional Neural Networks

Stability of Neural Networks Dependent on Time Series in Anime Image Recognition

Wing Interferential Patterns (WIPs) and machine learning for the classification of some Aedes species of medical interest

Contact Info

Product

Resources

About