Tomato Fruit Counting Dataset

The current agricultural industry faces challenges of inefficiency and lack of accuracy in fruit counting, especially in large-scale cultivation. Manual counting is not only time-consuming but also prone to errors. Most existing solutions rely on traditional image processing techniques, which cannot adapt to complex environmental changes and the diversity of fruit types. Therefore, the development of an efficient and accurate tomato fruit counting dataset aims to improve counting precision and efficiency through deep learning technology. This dataset contains images of tomatoes in various scenarios, which, through effective data annotation and processing, aids researchers and developers in training for object detection and image recognition. Data collection is conducted using drones and high-resolution cameras in greenhouse environments, ensuring high-quality image acquisition. For quality control, a combination of multiple rounds of annotation and expert review ensures data annotation accuracy and consistency. Data is stored in JPEG format, organized by image ID for quick retrieval and access. The dataset has a clear structure, facilitating subsequent model training and testing.