Apple Vision

In recent years, there has been a growing interest in using deep learning methods for Apple maturity detection. The main advantage of using these methods is their ability to learn and detect complex patterns in images, making them highly effective at distinguishing between different ripeness levels of apples. One such deep learning approach is YOLOv8, an object detection algorithm that uses a single neural network to detect objects in real time. The process of apple maturity detection using CNN and YOLOv8 involves collecting a large dataset of images of apples at different stages of ripeness, with labels indicating their maturity level. The dataset is then pre-processed, including image resizing, normalization, and augmentation. The CNN and YOLOv8 models are then trained on the pre-processed dataset, and their accuracy is evaluated using validation and testing datasets.

Once the model is trained and validated, it can be deployed in real-world scenarios, such as fruit sorting machines or mobile apps for consumers. By automating the Apple maturity detection process, growers and distributors can save time and reduce errors, resulting in improved efficiency and profitability.

Minimum Requirements:

1. Laptop or Computer.
2. Internet connection.
3. Web browser.

Recommended Requirements:

1. Laptop or Computer.
2. Modern web browser with JavaScript enabled, Google Chrome recommended.
3. Stable internet connection with a speed of at least 10 Mbps or higher.

Dataset:

Images are taken from:

1. Fruits 360 Dataset available on Kaggle. It’s a dataset of images containing fruits and vegetables containing 90483 images
2. Test Images are taken from Getty Images and from the web

References:

1. Raja Hamza and Mohamed Chtourou, “Apple Ripeness Estimation using Artificial Neural Network,” in International Conference on High-Performance Computing & Simulation, 2005
2. Raymond Erz Saragih and Andi W.R. Emanuel, “Banana Ripeness Classification Based on Deep Learning using Convolutional Neural Network,” in 3rd east Indonesia Conference on Computer and IT, 2021
3. Kaveri Kangune, Dr. Vrushali Kulkarni, and Prof. Pranali Kosamkar, “Grapes Ripeness Estimation using Convolutional Neural Network and Support Vector Machine,” in Global Conference for Advancement in Technology (GCAT) Bangalore, India, 2019
4. Brahim Benmouna1 · Ginés Garcia‑Mateos1 · Sajad Sabzi2 · Ruben Fernandez‑Beltran1 · Dolores Parras‑Burgos3 · José Miguel Molina‑Martinez4 “Convolutional Neural Networks for Estimating the Ripening State of Fuji Apples Using Visible and Near‑Infrared Spectroscopy” in Food and Bioprocess Technology (2022) 15:2226–2236”.
5. Yanfei Li, Xianying Feng, Yandong Liu & Xingchang Han “Apple quality identification and classification by image processing based on convolutional neural networks” in Scientific Reports | (2021) 11:16618.
6. Real-time multi-object tracking and segmentation using Yolov8 with DeepOCSORT and OSNet, https://github.com/mikel-brostrom/yolov8_tracking (Feb 10, 2023)
7. J.Jayanth, Mansa Mahadevaswamy & M.Shivakumar “Fruit Quality Identification and Classification by Convolutional Neural Network”. SN COMPUT. SCI. 4, 220 (2023). https://doi.org/10.1007/s42979-022-01527-w