Dataset of banana leaves and stem images for object detection, classification and segmentation: A case of Tanzania

Mduma, Neema; Leo, Judith

doi:10.1016/j.dib.2023.109322

Cited by 3 publications

(3 citation statements)

References 2 publications

(2 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…1 Firstly, banana disease images were fed to DL based feature extractor capable of capturing spatial correlation among neighboring pixels in the receptive area defined by the convolution kernel size by overriding directional information inconsideration of distance between the pixels [23] for subsequent steps of framework. Subsequently, from the extracted features via EfficienetB0 [24] for banana disease datasets [25], [26] for disease detection. To address the model explainability we adopt GradCAM visualization for understanding [22].…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Explainable AI Based framework for Banana Disease Detection

muaad

et al. 2024

Preprint

View full text Add to dashboard Cite

Due to widespread usage of banana as a staple food crop and susceptibility to numerous illnesses. Bananas require sophisticated detection techniques to support sustainable agricultural practices. Bananas are particularly susceptible to various stem and leaf spot diseases, resulting in significant economic losses within the banana cultivation sector. In this paper, a new XAI framework for banana disease detection and classification is introduced. Our framework uses state-of-the-art AI methods to analyze photos of banana plants. With great precision, it can detect a variety of illnesses like Cordana, Black Sigatoka, Pestalotiopsis, and fusarium wilt. The outcomes show that the framework performs better than current techniques in precisely identifying and categorizing banana diseases. The research employed a Convolutional Neural Networks (CNNs) to detect diseases in banana plants using RGB images of banana leaves. We used pre-trained model called EfficientnetB0 model to evaluate using two datasets BLSD and BDT. For BLSD, the model achieved an accuracy of 99.22%. Next for BDT, on the other hand, demonstrated improved performance with an accuracy of 99.63%.

show abstract

Section: Methodsmentioning

confidence: 99%

“…The XAI framework has preprocessing step where diseased banana images [26] resized to the model's fixed dimensions of 224x224. After resizing of images, we divided the dataset into three segments: 70% for training, 20% for testing, and 10% for validation.…”

Section: B Data Preprocessingmentioning

confidence: 99%

Explainable AI Based framework for Banana Disease Detection

muaad

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…We use public data collected by [14]. The dataset contains RGB images of banana leaves and stems representing healthy banana trees, black sigatoka disease, and fusarium wilt race 1 disease.…”

Section: Datasetmentioning

confidence: 99%

Deep learning model for detection and classification of banana diseases based on leaf images

Nassor,

Mushthofa,

Priandana

2024

IOP Conf. Ser.: Earth Environ. Sci.

View full text Add to dashboard Cite

Fungal diseases are among the main reasons for low productivity in banana farming. Early detection of fungal diseases is essential. One possible approach is using machine vision. Due to its high accuracy, deep learning is the most widely used algorithm in machine vision for many solutions. Its ability to model the data into multiple levels of abstraction makes it suitable for many agricultural solutions. However, deep learning requires a high computational resource, challenging many agricultural solutions implemented on low-computing devices. This study proposes lightweight deep-learning algorithms for detecting and classifying banana diseases based on leaf images. The study used a dataset of images representing three classes: black Sigatoka disease, fusarium wilt race 1 disease, and healthy tree. The algorithms used are mobileNetv2, mobileNetv3-small, shuffleNetv2, and squeezeNet. The results showed that squeezeNet outperforms all other models with 97.12% accuracy, 97.14% precision, 97.1% recall, and 97.12% f1-score. MobileNetv3-small results in the heaviest model, which is 14 MB, but it has the shortest training time of 2.465 minutes. MobileNetv2 results in the lightest model, 2.51 MB, while squeezeNet has the longest training time, 14.76 minutes. Overall, the lightweight deep learning algorithms performed well, and this method can be used for other banana diseases and abnormalities.

show abstract