Precise Agriculture: Effective Deep Learning Strategies to Detect Pest Insects

Butera, Luca; Ferrante, Alberto; Jermini, Mauro; Prevostini, Mauro; Alippi, Cesare

doi:10.1109/jas.2021.1004317

Cited by 38 publications

(32 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Articles follow this approach [7]- [10], and each of the previous articles uses different methods. However, the fact that they all focus on well-known insects is what unites them (e.g., spiders, mosquitoes, etc).…”

Section: Web Sourcesmentioning

confidence: 99%

“…The most well-known collection methods involve the use of search engines like Google [11] and Bing [12] or the use of websites like Flickr [13] as the standard gathering techniques. While Butera et al [7] employ all three techniques, other authors [8], [9] use Google and Flickr, and some authors [10] do not mention the techniques employed.…”

Section: Web Sourcesmentioning

confidence: 99%

“…-Some articles only detect one insect or one class: [9], [19], [23], [30], [41], [42], [44], [50], [52], [82], [90], -Others detect two classes: [8], [14], [26], [46], -Other papers detect 3 classes: [7], [20], [27], [28], [32], [48], [51], -Other articles detect 4 classes: [17], and [89], -Some of the studied papers detect 5 classes: [35], [10], and [39], -There are some papers that detect 6 classes: [15], [16], [18], [25], [34], -Other papers detect from 7 to 9 classes: [24] (7 classes), [31], [37], and [85] Since there are fewer studies as the number of insects increases, 66% of all articles focus on insects with a population of one to six.…”

Section: Existing Classification Techniquesmentioning

confidence: 99%

See 2 more Smart Citations

A Systematic Literature Review on Insect Detection in Images

Souifi¹,

Mdhaffar²,

Rodriguez³

et al. 2022

Signal, Image Processing and Embedded Systems Trends

View full text Add to dashboard Cite

Due to the advancements of deep learning (DL), particularly in the areas of visual object detection and convolutional neural networks (CNN), insect detecetion in images has received a lot of attention from the research community in the last few years. This paper presents a systematic review of the literature on the topic of insect detection as a case of object detection in images. It covers 50 research papers on the subject and responds to three research questions: i) type of dataset used; ii) detection technique used; iii) insect location. The paper also provides a summary of existing methods used for insect detection.

show abstract

Section: Web Sourcesmentioning

confidence: 99%

Section: Web Sourcesmentioning

confidence: 99%

Section: Existing Classification Techniquesmentioning

confidence: 99%

See 1 more Smart Citation

A Systematic Literature Review on Insect Detection in Images

Souifi¹,

Mdhaffar²,

Rodriguez³

et al. 2022

Signal, Image Processing and Embedded Systems Trends

View full text Add to dashboard Cite

show abstract

“…It is increasingly being used in various industrial applications for decision support to increase productivity, reduce errors, and reduce costs. Deep learning techniques perform better than traditional artificial intelligence techniques in terms of decision accuracy and reliability [ 6 ]. Deep convolutional neural networks (DCNN) are a class of supervised deep learning techniques.…”

Section: Introductionmentioning

confidence: 99%

An Improved Deep Residual Convolutional Neural Network for Plant Leaf Disease Detection

Pandian

Kanchanadevi

Rajalakshmi

et al. 2022

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

In this research, we proposed a novel deep residual convolutional neural network with 197 layers (ResNet197) for the detection of various plant leaf diseases. Six blocks of layers were used to develop ResNet197. ResNet197 was trained and tested using a combined plant leaf disease image dataset. Scaling, cropping, flipping, padding, rotation, affine transformation, saturation, and hue transformation techniques were used to create the augmentation data of the plant leaf disease image dataset. The dataset consisted of 103 diseased and healthy image classes of 22 plants and 154,500 images of healthy and diseased plant leaves. The evolutionary search technique was used to optimise the layers and hyperparameter values of ResNet197. ResNet197 was trained on the combined plant leaf disease image dataset using a graphics processing unit (GPU) environment for 1000 epochs. It produced a 99.58 percentage average classification accuracy on the test dataset. The experimental results were superior to existing ResNet architectures and recent transfer learning techniques.

show abstract

“…Deep learning is a type of artificial intelligence technique extended from artificial neural networks that imitates how humans make intelligent decisions by acquiring new knowledge 4 . The filters and layers in deep neural networks extract accurate features from the training data and perform better than traditional artificial intelligence techniques in terms of decision accuracy and reliability 5 . Deep convolutional neural network (DCNN) is a class of supervised deep learning techniques that are most successful when applied to image classification and object detection tasks.…”

Section: Introductionmentioning

confidence: 99%

An improved deep convolutional neural network for detecting plant leaf diseases

Pandian

Kanchanadevi

2022

Concurrency and Computation

View full text Add to dashboard Cite

This research proposes a novel dense convolutional neural network with five dense blocks (5DB-DenseConvNet) for the detection of plant leaf diseases. Five dense blocks and four transition layers were used to develop the 5DB-DenseConvNet. The 5DB-DenseConvNet was trained and tested using a plant leaf disease image dataset. Such advanced data augmentation techniques as color augmentations, geometric transformations, neural style transfer, principal component analysis, color augmentation, random erasing, and super-resolution generative adversarial network were used to increase the dataset size. The dataset contains 97 diseased and healthy image classes of 21 plants and comprises 145,500 images of healthy and diseased plant leaves. The Bayesian searching technique was used to optimize the hyperparameter values of 5DB-DenseConvNet. Compression factor was introduced in the transition layer to improve the compactness of the 5DB-DenseConvNet. The 5DB-DenseConvNet was trained on a graphics processing unit environment for 1000 epochs. The trained 5DB-DenseConvNet performed at an average classification accuracy of 99.36% and at an average precision of 98.83% on the test dataset. The experimental results demonstrate that the performance and reliability of the proposed 5DB-DenseConvNet on plant disease detection are superior to prior transfer learning techniques.

show abstract

Precise Agriculture: Effective Deep Learning Strategies to Detect Pest Insects

Cited by 38 publications

References 31 publications

A Systematic Literature Review on Insect Detection in Images

A Systematic Literature Review on Insect Detection in Images

An Improved Deep Residual Convolutional Neural Network for Plant Leaf Disease Detection

An improved deep convolutional neural network for detecting plant leaf diseases

Contact Info

Product

Resources

About