Enhancing 2D-3D facial recognition accuracy of truncated-hiden faces using fused multi-model biometric deep features DOI
Imen Labiadh, Larbi Boubchir, Hassene Seddik

и другие.

Multimedia Tools and Applications, Год журнала: 2024, Номер unknown

Опубликована: Май 9, 2024

Язык: Английский

Apnet: Lightweight network for apricot tree disease and pest detection in real-world complex backgrounds DOI Creative Commons
Minglang Li, Zhiyong Tao,

Wentao Yan

и другие.

Plant Methods, Год журнала: 2025, Номер 21(1)

Опубликована: Янв. 9, 2025

Язык: Английский

Процитировано

4

PND-Net: plant nutrition deficiency and disease classification using graph convolutional network DOI Creative Commons
Asish Bera, Debotosh Bhattacharjee, Ondřej Krejcar

и другие.

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Июль 5, 2024

Crop yield production could be enhanced for agricultural growth if various plant nutrition deficiencies, and diseases are identified detected at early stages. Hence, continuous health monitoring of is very crucial handling stress. The deep learning methods have proven its superior performances in the automated detection deficiencies from visual symptoms leaves. This article proposes a new method disease classification using graph convolutional network (GNN), added upon base neural (CNN). Sometimes, global feature descriptor might fail to capture vital region diseased leaf, which causes inaccurate disease. To address this issue, regional holistic aggregation. In work, region-based summarization multi-scales explored spatial pyramidal pooling discriminative representation. Furthermore, GCN developed capacitate finer details classifying insufficiency nutrients. proposed method, called Plant Nutrition Deficiency Disease Network (PND-Net), has been evaluated on two public datasets deficiency, four backbone CNNs. best PND-Net as follows: (a) 90.00% Banana 90.54% Coffee deficiency; (b) 96.18% Potato 84.30% PlantDoc Xception backbone. additional experiments carried out generalization, achieved state-of-the-art datasets, namely Breast Cancer Histopathology Image Classification (BreakHis 40

Язык: Английский

Процитировано

17

Small size CNN (CAS-CNN), and modified MobileNetV2 (CAS-MODMOBNET) to identify cashew nut and fruit diseases DOI
Kamini G. Panchbhai, Madhusudan G. Lanjewar,

Vishant V. Malik

и другие.

Multimedia Tools and Applications, Год журнала: 2024, Номер unknown

Опубликована: Апрель 2, 2024

Язык: Английский

Процитировано

13

Transfer learning in agriculture: a review DOI Creative Commons
Md Ismail Hossen, Mohammad Awrangjeb, Shirui Pan

и другие.

Artificial Intelligence Review, Год журнала: 2025, Номер 58(4)

Опубликована: Янв. 25, 2025

Язык: Английский

Процитировано

1

Enhanced hybrid attention deep learning for avocado ripeness classification on resource constrained devices DOI Creative Commons
Sumitra Nuanmeesri

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Янв. 29, 2025

Attention mechanisms such as the Convolutional Block Module (CBAM) can help emphasize and refine most relevant feature maps color, texture, spots, wrinkle variations for avocado ripeness classification. However, CBAM lacks global context awareness, which may prevent it from capturing long-range dependencies or patterns relationships between distant regions in image. Further, more complex neural networks improve model performance but at cost of increasing number layers train parameters, not be suitable resource constrained devices. This paper presents Hybrid Neural Network (HACNN) classifying on It aims to perform local enhancement capture relationships, leading a comprehensive extraction by combining attention modules models. The proposed HACNN combines transfer learning with hybrid mechanisms, including Spatial, Channel, Self-Attention Modules, effectively intricate features fourteen thousand images. Extensive experiments demonstrate that EfficienctNet-B3 significantly outperforms conventional models regarding accuracy 96.18%, 92.64%, 91.25% train, validation, test models, respectively. In addition, this consumed 59.81 MB memory an average inference time 280.67 ms TensorFlow Lite smartphone. Although ShuffleNetV1 (1.0x) consumes least resources, its testing is only 82.89%, insufficient practical applications. Thus, MobileNetV3 Large exciting option has 91.04%, usage 26.52 MB, 86.94 These findings indicated method enhances classification ensures feasibility implementation low-resource environments.

Язык: Английский

Процитировано

1

A hybrid deep learning model approach for automated detection and classification of cassava leaf diseases DOI Creative Commons

G. Sambasivam,

G. Prabu Kanna, Munesh Singh Chauhan

и другие.

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Фев. 27, 2025

Detecting cassava leaf disease is challenging because it hard to identify diseases accurately through visual inspection. Even trained agricultural experts may struggle diagnose the correctly which leads potential misjudgements. Traditional methods these are time-consuming, prone error, and require expert knowledge, making automated solutions highly preferred. This paper explores application of advanced deep learning techniques detect as well classify includes EfficientNet models, DenseNet169, Xception, MobileNetV2, ResNet Vgg19, InceptionV3, InceptionResNetV2. A dataset consisting around 36,000 labelled images leaves, afflicted by such Cassava Brown Streak Disease, Mosaic Green Mottle, Bacterial Blight, healthy was used train models. Further were pre-processed converting them into grayscale, reducing noise using Gaussian filter, obtaining region interest Otsu binarization, Distance transformation, Watershed technique followed employing contour-based feature selection enhance model performance. Models, after fine-tuned with ADAM optimizer computed that among tested hybrid (DenseNet169 + EfficientNetB0) had superior performance classification accuracy 89.94% while EfficientNetB0 highest values precision, recall, F1score 0.78 each. The novelty lies in its ability combine DenseNet169's reuse capability EfficientNetB0's computational efficiency, resulting improved scalability. These results highlight for accurate scalable diagnosis, laying foundation plant monitoring systems.

Язык: Английский

Процитировано

1

Potato Harvesting Prediction Using an Improved ResNet-59 Model DOI
Abdelaziz A. Abdelhamid, Amel Ali Alhussan,

Al-Seyday T. Qenawy

и другие.

Potato Research, Год журнала: 2024, Номер unknown

Опубликована: Авг. 1, 2024

Язык: Английский

Процитировано

9

An ensemble deep learning models approach using image analysis for cotton crop classification in AI-enabled smart agriculture DOI Creative Commons

Muhammad Farrukh Shahid,

Tariq Jamil Saifullah Khanzada, Muhammad Ahtisham Aslam

и другие.

Plant Methods, Год журнала: 2024, Номер 20(1)

Опубликована: Июль 14, 2024

Agriculture is one of the most crucial assets any country, as it brings prosperity by alleviating poverty, food shortages, unemployment, and economic instability. The entire process agriculture comprises many sectors, such crop cultivation, water irrigation, supply chain, more. During cultivation process, plant exposed to challenges, among which pesticide attacks disease in are main threats. Diseases affect yield production, affects country's economy. Over past decade, there have been significant advancements agriculture; nevertheless, a substantial portion yields continues be compromised diseases pests. Early detection prevention for successful management.

Язык: Английский

Процитировано

7

Maintaining Symmetry between Convolutional Neural Network Accuracy and Performance on an Edge TPU with a Focus on Transfer Learning Adjustments DOI Open Access
Christian DeLozier, Justin A. Blanco,

Ryan Rakvic

и другие.

Symmetry, Год журнала: 2024, Номер 16(1), С. 91 - 91

Опубликована: Янв. 11, 2024

Transfer learning has proven to be a valuable technique for deploying machine models on edge devices and embedded systems. By leveraging pre-trained fine-tuning them specific tasks, practitioners can effectively adapt existing the constraints requirements of their application. In process adapting an model, practitioner may make adjustments model architecture, including input layers, output intermediate layers. Practitioners must able understand whether modifications will symmetrical or asymmetrical with respect performance. this study, we examine effects these runtime energy performance processor performing inferences. Based our observations, recommendations how adjust convolutional neural networks during transfer maintain symmetry between accuracy its We observe that TPU is generally more efficient than CPU at inferences networks, continues outperform as depth width network increases. explore multiple strategies adjusting layers demonstrate important cliffs consider when modifying model.

Язык: Английский

Процитировано

5

Hybrid methods for detection of starch in adulterated turmeric from colour images DOI
Madhusudan G. Lanjewar,

Satyam S. Asolkar,

Jivan S. Parab

и другие.

Multimedia Tools and Applications, Год журнала: 2024, Номер 83(25), С. 65789 - 65814

Опубликована: Янв. 19, 2024

Язык: Английский

Процитировано

4