RoseLeafInsight: A High-Resolution Image Dataset for Rose Leaf Disease Recognition

RoseLeafInsight is a meticulously curated high-resolution image dataset designed for the classification and recognition of various rose leaf conditions using machine learning and computer vision techniques. This dataset includes four categories of rose leaves: Healthy, Black Spot, Insect Hole, and Yellow Mosaic Virus, providing a diverse set of images for disease detection and automated plant health monitoring. Each category is well-represented, ensuring a balanced dataset suitable for developing deep learning models for classification, segmentation, and disease detection tasks.

Dataset Composition:
The dataset consists of a total of 3,228 high-resolution images, distributed across the following categories:
1. Healthy Leaf: 1,686 images
2. Black Spot: 409 images
3. Insect Hole: 453 images
4. Yellow Mosaic Virus: 680 images

The extended dataset (by augmentation) consists of a total of 12,000 high-resolution images, distributed across the following categories:
1. Healthy Leaf: 3,000 images
2. Black Spot: 3,000 images
3. Insect Hole: 3,000 images
4. Yellow Mosaic Virus: 3,000 images

Geographical Location of Data Collection:
The rose leaf images were collected from two distinct locations in Bangladesh, ensuring diversity in environmental conditions and plant health variations:
1. Zailla, Singair, Manikganj
- Latitude: 23°47'46.11"N
- Longitude: 90°13'15.73"E

2. Golap Gram, Sadullapur-Komolapur, Road Birulia Bridge, Dhaka 1216
- Latitude: 23°50'6.108''N
- Longitude: 90°18'31.5108''E
These locations are known for their extensive rose cultivation, making them ideal for collecting a dataset that captures real-world variations in rose leaf health and disease conditions.

Preprocessing Details:
To enhance model performance and standardize input images, the following preprocessing steps were applied:
• Resizing: All images were resized to 3000 × 3000 pixels for uniformity.
• Background Removal: Unwanted backgrounds were eliminated to focus on leaf features.
• Brightness Enhancement: The brightness of each image was adjusted by a factor of 1.2 to improve visibility and contrast.

Potential Applications:
RoseLeafInsight is ideal for training and evaluating machine learning and deep learning models in various applications, including:
• Automated plant disease detection systems
• Smart agriculture and precision farming
• Image-based disease diagnosis for plant pathology research
• Transfer learning and fine-tuning deep learning models for plant health classification

This dataset provides a valuable resource for researchers, agronomists, and AI practitioners seeking to develop robust solutions for real-time rose leaf disease detection.