Qilin Watermelon Dataset

Introduction

The Qilin Watermelon dataset is a unique collection that explores the relationship between watermelon appearance, knock sound, and sweetness. It aims to facilitate research and analysis in non‑destructive watermelon quality assessment.

Dataset Description

The dataset contains two main parts:

1. wav files: Audio recordings of the knocking sound produced when a watermelon is tapped, capturing acoustic properties that may reveal internal structure and maturity.
2. jpg files: Images showing each watermelon’s external appearance, including color, texture, and shape.

In addition to the audio and visual data, the dataset includes sweetness measurements obtained with a refractometer. These quantitative values can be correlated with the knock sound and visual appearance.

Getting Started

To begin using the Qilin Watermelon dataset, follow these steps:

1. Download the dataset from IEEE DataPort: Qilin Watermelon Dataset Domestic users’ Baidu Netdisk link

2. Clone the repository containing code for processing the dataset:

   git clone https://github.com/crf0409/watermelon_eval.git
   

3. Update the file paths in the code to match the location where you downloaded the dataset.

4. Run the main script to start exploring the dataset:

   python main.py
   

Contact Information

If you have any questions, suggestions, or are interested in research collaboration related to the Qilin Watermelon dataset, please feel free to contact me at crf88967835@outlook.com. I am happy to discuss further and provide additional information.

License

The Qilin Watermelon dataset is released under the CC BY‑NC 4.0 license. See the LICENSE file for details.