Pistachio is a nutritious nut that has many uses in the food industry. Iran is one of its largest producers, and pistachio is considered as a strategic export product for this country. This product has a great variety, most of which are cultivated in Iran and taken to other countries. Pistachios are sorted based on the shape of their shell into two categories: Open-mouth and Closed-mouth. The open-mouth pistachios are higher in price, value, and demand than the closed-mouth pistachios.
In the countries that are famous in pistachio production and exporting, there are companies that pack the picked pistachios from the trees and make them ready for exporting. As there are differences between the price and the demand of the open-mouth and closed-mouth pistachios, it is considerable for these companies to know precisely how much of these two kinds of pistachios exist in each packed package. We have introduced and shared a new dataset of pistachios. Pistachio is known as Pesteh in Iran, and that is why we called our dataset Pesteh-Set.
source paper: arxiv.org/abs/2005.03990
Published paper: 10.1007/s42044-021-00090-6
The whole dataset can be downloaded at https://www.kaggle.com/mohammadrahimzadeh/pistachios
The training and counting code is shared at https://github.com/mr7495/Pistachio-Counting
Pistachios Assortment
Pesteh-Set is made of two parts. The first part includes 423 images with ground truth. We sorted the pistachios into two classes: Open-mouth and closed-mouth. The ground truth of the images is a CSV file that consists of the bounding boxes of the two classes of pistachios in the images. There are between 1 to 27 pistachios in each image, and 3927 pistachios totally. The second part includes 6 videos with a total length of 167 seconds and 561 moving pistachios.
We have selected some frames of the videos and labeled them with a self-developed program using OpenCV library on python language. The images of the dataset were resized to 1070 × 600 pixels to save computing costs. Some of the images of this dataset are presented in the figure below:
The videos of the dataset have been recorded by a cell-phone camera with 1920 × 1080 pixels resolution, five of these videos are recorded with 60 frames per second(fps) frame rate, and one other is recorded with 30 fps frame rate. The cell-phone was perched on the wall above the line that was transporting the pistachios. This line was designed somehow that the pistachios could roll on it. The reason the pistachios rolling is so important is that the open-mouth pistachios could appear on their backside where they look like closed-mouth pistachios, but the rolling cause them to show their open-mouth side when rolling. Next figure presents a view of how the dataset was recorded, and also the general schematic of our proposed method for remote counting the pistachios.
The General View of how Pesteh-Set was recoreded and our proposed way for counting the pistachios
The self-developed program for labeling the images along all the codes designed for preparing and visualzing the data is avaialbe on this repository. We hope that researchers could use them to make the Pistachio-Dataset larger.
The details of the Pesteh-Set are listed in the next table.
| Data | Open-Mouth Pistachios | Closed-Mouth Pistachios | ALL the Pistachios |
|---|---|---|---|
| Video1 | 50 | 20 | 70 |
| Video2 | 60 | 20 | 80 |
| Video3 | 70 | 20 | 90 |
| Video4 | 90 | 20 | 110 |
| Video5 | 100 | 20 | 120 |
| Video6 | 39 | 52 | 91 |
| All of the Videos | 409 | 152 | 561 |
| All the 423 Labeld Images | 1993 | 1934 | 3927 |
Clone this repository by !git clone https://github.com/mr7495/Pesteh-Set.git
You can use !python frame_generator.py --video_path="Path to the Video" --save_folder="Path to write the frames" --height="height of the frame" --width="width of the frame" to extract the frames of the videos.
You can extract more frames from the videos in our dataset and use the Label_maker module to label more images and increase the number of images in the dataset.
Label Maker has been developed for making the ground truth of the images. In this case, as we have two classes of pistachios, the label maker is set to label two classes of objects.
The output of this program is a set of CSV files that each belongs to an image. One example of a row in a generated CSV file is :
image_name, x1, y1, x2, y2, class_num
This format is the format that RetinaNet accepts as the CSV annotation.
This program can be ran by :
!python Label_maker.py --images_path="path to the images folder" --save_path="The path to save csv files"
For each image, draw the box around the object and press enter, then press 0 or 1. 0 is referred to class 1, and 1 is referred to class2. If you press any other keys after pressing enter, the selected box would be rejected. If you want to go to the next image, select a random box, press enter, then press n. if you want to quit, select a random box, press enter, then press e
The visualization.py presents the code for plotting the annotation on the images. Run as:
!python visualization.py --annotation_path="Path to annotation.csv" --images_path="Path to the dataset images" --save_path="Path to write the plotted images"
Ground Truth of Pesteh-Set is available in the annotation.csv.
The Pesteh-Set images are available in the pesteh_set_images folder.
The Videos of Pesteh-Set is shared on Google Drive and Mendeley:
https://drive.google.com/drive/folders/1ojq1maRM82o3Ee4cOUelNFKqTXda_ovs?usp=sharing
https://data.mendeley.com/datasets/45hs5ht6k3/4
Also, the whole dataset can be downloaded at https://www.kaggle.com/mohammadrahimzadeh/pistachios
The paper is published at:
https://arxiv.org/abs/2005.03990
https://doi.org/10.1007/s42044-021-00090-6
If you find our work useful, please cite it by:
@article{rahimzadeh2021detecting,
title={Detecting and counting pistachios based on deep learning},
author={Rahimzadeh, Mohammad and Attar, Abolfazl},
journal={Iran Journal of Computer Science},
pages={1--13},
year={2021},
publisher={Springer}
}
You can contact the developer by this email: mr7495@yahoo.com