A Comprehensive Hog Plum Leaf Disease Dataset for Enhanced Detection and Classification.

Data in Brief, Journal Year: 2025, Volume and Issue: 59, P. 111311 - 111311

Published: Jan. 22, 2025

A comprehensive Hog plum leaf disease dataset is greatly needed for agricultural research, precision agriculture, and efficient management of disease. It will find applications toward the formulation machine learning models early detection classification disease, thus reducing dependency on manual inspections timely interventions. Such a provides benchmark training testing algorithms, further enhancing automated monitoring systems decision-support tools in sustainable agriculture. enables better crop management, less use chemicals, more focused agronomical practices. This contribute to global research being carried out advancement disease-resistant plant strategy development practices productivity along with sustainability. These images have been collected from different regions Bangladesh. In this work, two classes were used: 'Healthy' 'Insect hole', representing stages progression. The augmentation techniques that involve flipping, rotating, scaling, translating, cropping, adding noise, adjusting brightness, contrast, scaling expanded 3782 20,000 images. formed very robust deep sets, hence

Language: Английский

---

Dataset of Seven Tropical Flower Species from Bangladesh: A Resource for Classification and Ecological Studies

Data in Brief, Journal Year: 2025, Volume and Issue: 59, P. 111374 - 111374

Published: Feb. 8, 2025

We present the Tropical Flower Dataset: Seven Species from Bangladesh, a collection of 4,319 high-quality images containing seven tropical flower species: Rose (827), Bougainvillea (580), Marigold (717), Hibiscus (548), Crown Thorn (583), Jungle Geranium (698), and Madagascar Periwinkle (366). All were taken using Redmi Note 11 smartphone at different locations within Dhaka Division. The dataset is beneficial for classification ecological purposes, encompassing diverse growth stages distinct lighting conditions. Future work will focus on improving generalization through data augmentation fine-tuning, aiming to enhance automated species recognition monitoring, biodiversity studies, botany education.

Language: Английский

---

UDCAD-DFL-DL: A Unique Dataset for Classifying and Detecting Agricultural Diseases in Dragon Fruits and Leaves

Data in Brief, Journal Year: 2025, Volume and Issue: 59, P. 111411 - 111411

Published: Feb. 19, 2025

Language: Английский

---

Enhancing Agricultural Diagnostics: Advanced Training of Pre-Trained CNN Models for Paddy Leaf Disease Detection

Machine Learning Research, Journal Year: 2025, Volume and Issue: 10(1), P. 1 - 13

Published: March 31, 2025

Timely and precise identification of foliar diseases is essential in contemporary agriculture to avert crop loss, enhance productivity, guarantee food security. Paddy, being one the most extensively farmed consumed staple crops globally, especially vulnerable several leaf that can markedly diminish yield. Conventional illness detection techniques, which depend significantly on manual observation expert evaluation, are frequently time-consuming, labor-intensive, susceptible discrepancies. These constraints need implementation automated efficient disease technologies. This research investigates utilization a pre-trained EfficientNetB3 convolutional neural network for categorization paddy diseases. The model was trained assessed rich diverse dataset comprising annotated pictures healthy sick leaves. performance evaluation included conventional classification criteria like as accuracy, precision, recall, F1-score ensure comprehensive assessment model's efficacy. exhibited exceptional performance, with an overall accuracy 96% prevalent elevated signifies proficiency generalizing effectively across categories imaging settings. findings underscore capability deep learning computer vision methodologies revolutionize agricultural operations by offering scalable, dependable, instantaneous solutions identification. suggested approach facilitates early diagnosis, aiding farmers agronomists executing timely treatments, hence minimizing loss enhancing production. Moreover, incorporation AI-driven technologies into current frameworks fosters sustainable farming strengthens resilience production systems. highlights significant influence artificial intelligence precision establishes basis additional investigation intelligent monitoring

Language: Английский

---

IDBGL: A Unique Image Dataset of Black Gram (Vigna mungo) Leaves for Disease Detection and Classification

Data in Brief, Journal Year: 2025, Volume and Issue: 59, P. 111347 - 111347

Published: Jan. 29, 2025

Black gram (Vigna mungo) is considered one of the most important pulse crops cultivated in Bangladesh because it a vital source nutrition and potential for raising good income. It those plants where leaves are affected by diseases. We observed that were diseased fields, we had difficulty collecting healthy samples. The crop different diseases attacking leaf tissues, causing heavy yield loss. can apply deep learning models to recognize their early stages timely interference. Diseases could be detected with automation process, from which much enhancement management black possible. Our purpose create unique dataset Bangladesh's Gram help global researchers build learning-automated system detection classification will assist farmers more awareness among agricultural stakeholders. original 4,038 images was collected Sirajganj Solonga regions Bangladesh. has five classes: Healthy, Cercospora Leaf Spot, Insect, Crinkle, Yellow Mosaic. This improve disease Grams developing effective computational applying advanced machine techniques.

Language: Английский

---

Deep Object Occlusion Relationship Detection Based on Associative Embedding Clustering

Technologies, Journal Year: 2025, Volume and Issue: 13(4), P. 143 - 143

Published: April 4, 2025

Visual relationship detection is crucial for understanding scenes depicted in images when aiming to detect objects within the image and recognize visual relationships between each pair of objects. Nevertheless, profound occlusion, as a typical existing constituting pivotal semantic feature, has regrettably been subjected insufficient scrutiny. To address this issue, we propose pioneering approach termed DOORD-AEC, which specifically designed detecting occlusion spatial among targets. DOORD-AEC introduces associative embedding clustering supervise convolutional neural network with two branches, enabling it take an input produce triplet set representing relationships. The learns simultaneously identify all targets occlusions that make up group them together using clustering. Additionally, contribute KORD dataset, novel challenging dataset We demonstrate effectiveness our method dataset.

Language: Английский

---