Effective Training of Convolutional Neural Networks for Insect Image Recognition

Martineau, Maxime; Raveaux, Romain; Chatelain, Clément; Conte, Donatello; Venturini, Gilles

doi:10.1007/978-3-030-01449-0_36

Cited by 8 publications

(2 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This transfer learning approach greatly expedites the training process and has previously achieved high accuracy in tests on the iNaturalist dataset of citizen science records (e.g. Cui, Song, Sun, Howard, & Belongie, ) and for the identification of insects (Martineau, Raveaux, Chatelain, Conte, & Venturini, ). To repurpose the model, we replaced the imageNet classification layer with new layers and trained the model on our dataset.…”

Section: Methodsmentioning

confidence: 99%

Thinking like a naturalist: Enhancing computer vision of citizen science images by harnessing contextual data

2020

View full text Add to dashboard Cite

1. The accurate identification of species in images submitted by citizen scientists is currently a bottleneck for many data uses. Machine learning tools offer the potential to provide rapid, objective and scalable species identification for the benefit of many aspects of ecological science. Currently, most approaches only make use of image pixel data for classification. However, an experienced naturalist would also use a wide variety of contextual information such as the location and date of recording.2. Here, we examine the automated identification of ladybird (Coccinellidae) records from the British Isles submitted to the UK Ladybird Survey, a volunteer-led mass participation recording scheme. Each image is associated with metadata; a date, location and recorder ID, which can be cross-referenced with other data sources to determine local weather at the time of recording, habitat types and the experience of the observer. We built multi-input neural network models that synthesize metadata and images to identify records to species level.3. We show that machine learning models can effectively harness contextual information to improve the interpretation of images. Against an image-only baseline of 48.2%, we observe a 9.1 percentage-point improvement in top-1 accuracy with a multi-input model compared to only a 3.6% increase when using an ensemble of image and metadata models. This suggests that contextual data are being used to interpret an image, beyond just providing a prior expectation. We show that our neural network models appear to be utilizing similar pieces of evidence as human naturalists to make identifications. 4. Metadata is a key tool for human naturalists. We show it can also be harnessed by computer vision systems. Contextualization offers considerable extra information, particularly for challenging species, even within small and relatively homogeneous areas such as the British Isles. Although complex relationships between disparate sources of information can be profitably interpreted by simple neural network architectures, there is likely considerable room for further progress. Contextualizing images has the potential to lead to a step change in the accuracy of automated S U PP O RTI N G I N FO R M ATI O NAdditional supporting information may be found online in the Supporting Information section.

show abstract

Section: Methodsmentioning

confidence: 99%

Thinking like a naturalist: Enhancing computer vision of citizen science images by harnessing contextual data

2020

View full text Add to dashboard Cite

show abstract

“…The two models were identical, with the only difference being the training approach applied. One of the MobileNetV2 models was trained using the transfer learning technique, which involves ne-tuning the parameters of a network that has been pre-trained on a large dataset (Martineau et al 2018). The other model was trained without transfer learning.…”

Section: Training Of the Convolutional Neural Networkmentioning

confidence: 99%

A novel non-destructive detection approach for seed cotton lint percentage by using deep learning

Geng,

Yan,

et al. 2024

Preprint

View full text Add to dashboard Cite

Background The lint percentage of seed cotton is one the most important parameters in evaluation the seed cotton quality, which affects the price of the seed cotton during the purchase and sale. The traditional method of measuring lint percentage is labor-intensive and time-consuming, and thus there is a need for an efficient and accurate method. In recent years, classification-based machine learning and computer vision have shown promise in solving various classification tasks. Results In this study, we propose a new approach for detecting lint percentage using MobileNetV2 and transfer learning. The model is deployed on the Lint Percentage detection instrument, which can rapidly and accurately determine the lint percentage of seed cotton. We evaluated the performance of the proposed approach using a dataset of 66924 seed cotton images from different regions of China. The results from the experiments showed that the model achieved an average accuracy of 98.43% in classification with an average precision of 94.97%, an average recall of 95.26%, and an average F1-score of 95.20%. Furthermore, the proposed classification model also achieved an average ac-curacy of 97.22% in calculating the lint percentage, showing no significant difference from the performance of experts (independent-samples t test, t = 0.019, p = 0.860). Conclusions This study demonstrates the effectiveness of the MobileNetV2 model and transfer learning in calculating the lint percentage of seed cotton. The proposed approach is a promising alternative to the traditional method, offering a rapid and accurate solution for the industry.

show abstract

Termite Pest Identification Method Based on Deep Convolution Neural Networks

Huang

Liu

et al. 2021

Journal of Economic Entomology

View full text Add to dashboard Cite

Several species of drywood termites, subterranean termites, and fungus-growing termites cause extensive economic losses annually worldwide. Because no universal method is available for controlling all termites, correct species identification is crucial for termite management. Despite deep neural network technologies’ promising performance in pest recognition, a method for automatic termite recognition remains lacking. To develop an automated deep learning classifier for termite image recognition suitable for mobile applications, we used smartphones to acquire 18,000 original images each of four termite pest species: Kalotermitidae: Cryptotermes domesticus (Haviland); Rhinotermitidae: Coptotermes formosanus Shiraki and Reticulitermes flaviceps (Oshima); and Termitidae: Odontotermes formosanus (Shiraki). Each original image included multiple individuals, and we applied five image segmentation techniques for capturing individual termites. We used 24,000 individual-termite images (4 species × 2 castes × 3 groups × 1,000 images) for model development and testing. We implemented a termite classification system by using a deep learning–based model, MobileNetV2. Our models achieved high accuracy scores of 0.947, 0.946, and 0.929 for identifying soldiers, workers, and both castes, respectively, which is not significantly different from human expert performance. We further applied image augmentation techniques, including geometrical transformations and intensity transformations, to individual-termite images. The results revealed that the same classification accuracy can be achieved by using 1,000 augmented images derived from only 200 individual-termite images, thus facilitating further model development on the basis of many fewer original images. Our image-based identification system can enable the selection of termite control tools for pest management professionals or homeowners.

show abstract

Effective Training of Convolutional Neural Networks for Insect Image Recognition

Cited by 8 publications

References 15 publications

Thinking like a naturalist: Enhancing computer vision of citizen science images by harnessing contextual data

Thinking like a naturalist: Enhancing computer vision of citizen science images by harnessing contextual data

A novel non-destructive detection approach for seed cotton lint percentage by using deep learning

Termite Pest Identification Method Based on Deep Convolution Neural Networks

Contact Info

Product

Resources

About