This dataset, initially consisting of 10,154 high-resolution images of five fruit types—apple, banana, mango, orange, and grapes—has been expanded to over 81,000 using advanced augmentation techniques like rotation, scaling, and brightness adjustment. The images are classified into three key categories: fresh, rotten, and formalin-mixed. This dataset offers a unique opportunity for researchers in computer vision, agriculture, and food safety to develop machine learning and deep learning models for tasks such as real-time fruit quality assessment, contamination detection, and automating food inspection processes. Its extensive and diverse collection makes it a valuable resource for innovations in public health, export quality control, and agricultural productivity.

(1) Everyone seeks fresh and high-quality fruits, as fruits naturally degrade over time, leading to spoilage. It is estimated that roughly one-third of harvested fruits rot, causing significant financial loss. Additionally, the sale of fruits is impacted by consumer concerns over the health risks associated with consuming spoiled or chemically treated fruits. The manual classification of fresh and rotten fruits by individuals is inefficient, particularly for farmers, sellers, and the fruit processing industry. (2) In recent years, computer vision techniques have shown great promise in automating tasks like classification and detection of fresh and rotten fruits. (3) To aid in the development of computer vision-based algorithms, we present an extensive fruit dataset containing five fruit classes: Fresh Apple, Rotten Apple, Formalin-mixed Apple, Fresh Banana, Rotten Banana, Formalin-mixed Banana, Fresh Grape, Rotten Grape, Formalin-mixed Grape, Fresh Mango, Rotten Mango, Formalin-mixed Mango, Fresh Orange, Rotten Orange, and Formalin-mixed Orange. Classification into fresh, rotten, and formalin-treated categories was carried out with the assistance of agricultural experts.

Fresh Rotten Fruit

## Overview

Fresh Rotten Fruit is a dataset for object detection tasks - it contains FruitFreshness annotations for 1,336 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Fruits Fresh And Rotten

## Overview

Fruits Fresh And Rotten is a dataset for object detection tasks - it contains Bounding Box annotations for 13,599 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

About Dataset (strawberries, peaches, pomegranates) Photo requirements: 1-White background 2-.jpg 3- Image size 300*300 The number of photos required is 250 photos of each fruit when it is fresh and 250 photos of each Fruit Dataset for Classification when it is rotten. Total 1500 images

Diverse Collection With a diverse collection of Product images, the files provides an excellent foundation for developing and testing machine learning models designed for image recognition and allocation. Each image is captured under different lighting conditions and backgrounds, offering a realistic challenge for algorithms to overcome.

Real-World Applications The variability in the dataset ensures that models trained on it can generalize well to real-world scenarios, making them robust and reliable. The dataset includes common fruits such as apples, bananas, oranges, and strawberries, among others, allowing for comprehensive training and evaluation.

Industry Use Cases One of the significant advantages of using the Fruits Dataset for Classification is its applicability in various fields such as agriculture, retail, and the food industry. In agriculture, it can help automate the process of fruit sorting and grading, enhancing efficiency and reducing labor costs. In retail, it can be used to develop automated checkout systems that accurately identify fruits, streamlining the purchasing process.

Educational Value The dataset is also valuable for educational purposes, providing students and educators with a practical tool to learn and teach machine learning concepts. By working with this dataset, learners can gain hands-on experience in data preprocessing, model training, and evaluation.

Conclusion The Fruits Dataset for Classification is a versatile and indispensable resource for advancing the field of image classification. Its diverse and high-quality images, coupled with practical applications, make it a go-to dataset for researchers, developers, and educators aiming to improve and innovate in machine learning and computer vision.

This dataset is sourced from Kaggle.

Dataset

This dataset was created by Shadab Ahmad

Contents

(strawberries, peaches, pomegranates) Photo requirements: 1-White background 2-.jpg 3- Image size 300*300 The number of photos required is 250 photos of each fruit when it is fresh and 250 photos of each fruit when it is rotten. Total 1500 images

FruQ-DB is a dataset containing 11 varieties of fruits images. These images were created from YouTube fruit time-lapse videos (see link to the videos below). The FruQ-DB database consist of image frames from fruits such as banana, cucumber, grape, kaki, papaya, peach, pear, pepper, strawberry, tomatoes, and watermelon. A total number of 5647 preprocessed images with de-watermarked and resized (224x224). Three classes of fruit quality with number of image samples per class is summarized below:

Fresh (2182 images)

Mild (1364 images)

Rotten (2101 images)

Another datasets is uploaded FruQ-Multi is another folder containing each types of fruits and their classes.

Links to the YouTube videos for the different fruit time elapse are:

Avocado https://www.youtube.com/watch?v=FeQehUXZYPk

Banana https://www.youtube.com/watch?v=OmcXo9XC6Uc

Cucumber https://www.youtube.com/watch?v=UMnevucxOug

Kaki (persimmon) https://www.youtube.com/watch?v=xE0Pw7jeOBo

Papaya https://www.youtube.com/watch?v=M8scWymSp2Y

Peach https://www.youtube.com/watch?v=g9pf19wk0-E&t=367s

Pepper https://www.youtube.com/watch?v=H0Sd6Foaepk

Strawberry https://www.youtube.com/watch?v=UMnevucxOug

Tomato https://www.youtube.com/watch?v=6xEcoU1vAZk

Watermelon https://www.youtube.com/watch?v=S12zZhdOckc

Here are a few use cases for this project:

1. Smart Grocery Shopping: The model can be integrated into a smart shopping system that enables customers to automatically scan and identify the quality of apples and peaches they are buying. This provides a helpful tool in promoting qualitative shopping and reducing food waste.

2. Agri-Tech Quality Control: Farmers and agricultural technologists can utilize this model for determining the quality of their produce in real-time, enabling them to separate fresh fruits from the rotten ones effectively, thereby improving overall productivity.

3. Educational Tool in Schools: The model can be used as a comprehensive learning resource in schools to educate students about different types of fruits, their sizes, and freshness levels.

4. Food Rescue Operations: For NGOs and organizations involved in food rescue operations, this model can quickly identify and segregate edible fruits from the rotten ones in large quantities.

5. Food Processing Industry: In industries like juice making, this model can be used to ensure the quality of fruits used. The system can identify and separate the rotten fruits, helping ensure only fresh fruits are processed.

Dataset

This dataset was created by Sriram Reddy Kalluri

Contents

Context

This dataset contains images of 6 fruits and vegetables: apple, banana, bitter gourd, capsicum, orange, and tomato. The images of each fruit or vegetable are grouped into two categories: fresh and stale. The purpose behind the creation of this dataset is the development of a machine learning model to classify fruits and vegetables as fresh or stale. This feature is a part of our final year project titled ‘Food Aayush’. (Github Link)

Data Collection and Preprocessing

For collecting the images to create the dataset, images of the fruits and vegetables were captured daily using a mobile phone camera. Depending on the visual properties of the fruit or vegetable in each image and on the day when the image was captured, each image was labelled as fresh or stale. Additionally, videos of the fruits and vegetables were taken, and the frames of these videos were extracted to collect a large number of images conveniently. The machine learning model requires a 224 x 224-pixel image. So, the images were cropped to extract the center square of the image and resized in 512 x 512 pixels using a data pre-processing library in Keras. Frame Extraction

Data Augmentation: We used ImageDataGenerator library from Keras for augmentation. We on average created 20 augmentations per image which indeed improve our models accuracy. Data Augmentation

Acknowledgements

We would like to give credit to this dataset as we have obtained the images in some of the classes from here. Dataset

Inspiration

Our BE final year project, titled ‘Food Aayush’, is an application that can be used for the classification of fruits and vegetables as fresh or stale, the classification of cooking oils into different rancidity levels, and the analysis of various parameters related to the nutritional value of food and people’s dietary intake. We have trained a machine learning model for the classification of fruits and vegetables. This dataset was created for training the machine learning model. Your data will be in front of the world's largest data science community. What questions do you want to see answered?

Apples are one of the most productive fruits in the world, in addition to their nutritional and health advantages for humans. Even with the continuous development of AI in agriculture in general and apples in particular, automated systems continue to encounter challenges identifying rotten fruit and variations within the same apple category, as well as similarity in type, color, and shape of different fruit varieties. These issues, in addition to apple diseases, substantially impact the economy, productivity, and marketing quality. In this paper, we first provide a novel comprehensive collection named Apple Fruit Varieties Collection (AFVC) with 29,750 images through 85 classes. Second, we distinguish fresh and rotten apples with Apple Fruit Quality Categorization (AFQC), which has 2,320 photos. Third, an Apple Diseases Extensive Collection (ADEC), comprised of 2,976 images with seven classes, was offered. Fourth, following the state of the art, we develop an Optimized Apple Orchard Model (OAOM) with a new loss function named measured focal cross-entropy (MFCE), which assists in improving the proposed model’s efficiency. The proposed OAOM gives the highest performance for apple varieties identification with AFVC; accuracy was 93.85%. For the apples rotten recognition with AFQC, accuracy was 98.28%. For the identification of the diseases via ADEC, it was 99.66%. OAOM works with high efficiency and outperforms the baselines. The suggested technique boosts apple system automation with numerous duties and outstanding effectiveness. This research benefits the growth of apple’s robotic vision, development policies, automatic sorting systems, and decision-making enhancement.

Rotten Fresh Fruitclassification

## Overview

Rotten Fresh Fruitclassification is a dataset for classification tasks - it contains Fruits annotations for 1,519 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

Fruit Maturity

## Overview

Fruit Maturity is a dataset for classification tasks - it contains Fruit annotations for 237 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

This dataset contains a total of 6.2k images of hand labeled everyday items and fruits which were augmented and this data with 62k images was created. The original data contains images that were taken live as well as images scraped from the internet. The data is stored in YOLO format.

The items included are - 'fresh apple', - 'fresh banana' - 'rotten apple' - 'rotten banana' - 'baidynath chyawanprash' - 'bajaj almond drop' - 'bournvita powder' - 'dabur chyawanprash' - 'dettol hand sanitizer' - 'dot and key facewash' - 'dove soap' - 'everyuth facewash' - 'exo safai scrub' - 'fogg spray' - 'garnier facewash' - 'godrej hand wash powder' - 'haldiram bhujia' - 'haldiram moongdal' - 'haldiram punjabi tadka' - 'head and shoulders shampoo' - 'joy body lotion' - 'kamasutra spray' - "Kellogg's chocos" - 'kurkure packet' - 'lakme sunscreen' - 'lays chips packet' - 'nivea body lotion' - 'odonil room freshner spray' - 'parachute jasmine oil' - 'patanjali aloevera gel' - 'saffola oats' - 'sensodyne toothpaste' - 'vaseline petroleum jelly' - 'vim bar'

Here are a few use cases for this project:

1. Grocery Store Quality Control: This computer vision model can be used by grocery stores and supermarkets to automatically check and sort fruits and vegetables according to their freshness levels. It can help to separate fresh produce from rotten ones, ensuring that consumers always have access to high quality produce.

2. Supply Chain Management: The model can help suppliers to visually inspect produce before it is sent to grocery stores or food processing companies. This will increase efficiency of the supply chain and reduce food waste.

3. Food Processing Plants: Food processing plants, particularly those that deal with canned fruits and vegetables, can use this model to sort and categorize fresh and rotten produce for different processing stages.

4. Agricultural Quality Inspection: Farms can employ this computer vision model to detect rotten fruits and vegetables right on the fields, allowing them to improve their quality control measures and avoid transporting rotten produce.

5. Personal Use - Health & Wellness Apps: This model can be incorporated into mobile applications, helping individuals at home to identify if their fruits and vegetables are still fresh and good for consumption. This will help people to consume healthier food and reduce food waste at home.

AG-Data is a comprehensive agricultural image classification dataset comprising 30,797 images across 46 distinct categories. We have split the dataset into a training set, a validation set, and a test set in a 6:2:2 ratio.The following is a detailed description of the AG-data dataset：1. The Sorghum dataset contains 3 categories: BroadLeafWeed (1441 images), class0_sorghum (1404 images), and class1_Grass (1467 images), with a total of 4312 images.2. The Banana dataset comprises 4 categories: cordana (400 images), pestalotiopsis (400 images), sigatoka (400 images), and healthy (400 images), totaling 1600 images.3. The sunflower dataset includes 4 categories: Downy mildew (470 images), Fresh leaf (515 images), Gray mold (398 images), and leaf scars (509 images), with a total of 1892 images.4. The mulberry dataset encompasses 10 categories, each with the following number of images: BlackAustralia (637), BlackOodTurkey (500), Buriram60 (345), ChiangMai60 (500), ChiangMaiBuriram60 (761), Kamphaengsaeng42 (500), RedKing (350), TaiwanMeacha (640), Taiwanstraberry (488), and WhiteKing (541), totaling 4497 images.5. The pomegranate dataset consists of 5 categories: Alternaria (886 images), Anthracnose (1166 images), Bacterial_Blight (966 images), Cercospora (631 images), and healthy (1450 images), with a total of 5099 images.6. The potatoleaf dataset has 7 categories: Bacteria (569 images), Fungi (748 images), Nematode (68 images), pest (611 images), Phytopthora (347 images), Virus (532 images), and healthy (201 images), totaling 3076 images.7. The RicePest dataset includes 10 categories, each with the following number of images: asiatic rice borer (498), brown plant hopper (346), paddy stem maggot (89), rice gall midge (217), rice leaf caterpillar (153), rice leaf roller (716), rice leaf hopper (244), rice water weevil (414), samll brown plant hopper (243), and yellow rice borer (236), totaling 3156 images.8. The cucumber dataset comprises 8 categories, each with 800 images, totaling 6400 images. The categories are: Anthracnose, Bacterial wilt, Belly Rot, Downy mildew, Fresh cucumber, Fresh leaf, Gummy Stem Blight, and Pythium Fruit Rot.

(1) Crop disease is a widespread problem in the productivity and quality of agricultural production. It adversely affects the quality of crops. The cucumber is a frequently grown creeping vine plant that has few calories but is high in water and several vital vitamins and minerals. Due to the non-biological circumstances, cucumber diseases will adversely harm the yield and quality of cucumber and cause heavy economic losses to farmers. The traditional diagnosis of crop diseases is often time-consuming, laborious, ineffective, and subjective.

(2) In the recent era, computer vision approaches are very promising for handling these kinds of classification and detection tasks.

(3) To develop machine vision-based algorithms, a major cucumber dataset is illustrated containing eight types of cucumber classes, namely Anthracnose, Bacterial Wilt, Belly Rot, Downy Mildew, Pythium Fruit Rot, Gummy Stem Blight, Fresh leaves, and Fresh cucumber. Cucumber disease classifications are done with the cooperation of an expert from an agricultural institute.

(4) A total of 1280 images of cucumbers are collected from real fields. Then from these original images, a total of 6400 augmented images are produced using flipping, shearing, zooming, and rotation techniques to increase the data number.

How to identify fresh and rotten fruits - aiding the elderly in identifying good and bad fruits easily if they have poor eyesight.

Here are a few use cases for this project:

1. Grocery Quality Control: This model can be used to identify the freshness status of fruits in supermarkets, groceries, and supply chains. By harnessing this, companies can improve the quality of products reaching their consumers and minimize incidental losses.

2. Smart Home Assistance: Integrated into smart home devices, it can help individuals identify the freshness of their stocked fruits, assisting in better management of home groceries and reducing food waste.

3. Fruit Production Monitoring: Agribusinesses could use this model to monitor the quality of their fruits during production and post-harvest handling. It can serve as an automated quality control tool to classify fresh fruits from the rotten ones.

4. Food Processing Industry: Companies that manufacture fruit-based products, such as juices (like Tropicana and Goodday), can utilize this model to verify the quality and freshness of the fruits used in their production process.

5. Unrelated Objects Detection: Though misplaced, the laundry detergent image signifies the model's scope extends to unrelated object detection as well. This makes it useful in scenarios where identifying out-of-place objects is critical, such as machine maintenance, security systems, etc.

Determining whether apples, bananas, and oranges are fresh or rotten