Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Cell Phone Image-Based Persian Rice Detection and Classification Using Deep Learning Techniques

Mahmood Saeedi kelishami, Amin Saeidi Kelishami, Sajjad Saeedi Kelishami

TL;DR

The paper tackles automatic Persian rice variety identification and bulk composition using consumer-grade images. It introduces a dual DL pipeline with ResNet50-based single-grain classification and U-Net-based bulk segmentation to support fine-grained and aggregate quality analysis. Results show 55% overall accuracy for grain-level classification and IoU of 93% for segmentation, enabling estimation of mixture proportions in bulk samples. The approach demonstrates the practicality of accessible smartphone imagery for rice quality assessment and offers directions for improving accuracy and extending to other foods.

Abstract

This study introduces an innovative approach to classifying various types of Persian rice using image-based deep learning techniques, highlighting the practical application of everyday technology in food categorization. Recognizing the diversity of Persian rice and its culinary significance, we leveraged the capabilities of convolutional neural networks (CNNs), specifically by fine-tuning a ResNet model for accurate identification of different rice varieties and employing a U-Net architecture for precise segmentation of rice grains in bulk images. This dual-methodology framework allows for both individual grain classification and comprehensive analysis of bulk rice samples, addressing two crucial aspects of rice quality assessment. Utilizing images captured with consumer-grade cell phones reflects a realistic scenario in which individuals can leverage this technology for assistance with grocery shopping and meal preparation. The dataset, comprising various rice types photographed under natural conditions without professional lighting or equipment, presents a challenging yet practical classification problem. Our findings demonstrate the feasibility of using non-professional images for food classification and the potential of deep learning models, like ResNet and U-Net, to adapt to the nuances of everyday objects and textures. This study contributes to the field by providing insights into the applicability of image-based deep learning in daily life, specifically for enhancing consumer experiences and knowledge in food selection. Furthermore, it opens avenues for extending this approach to other food categories and practical applications, emphasizing the role of accessible technology in bridging the gap between sophisticated computational methods and everyday tasks.

Cell Phone Image-Based Persian Rice Detection and Classification Using Deep Learning Techniques

TL;DR

The paper tackles automatic Persian rice variety identification and bulk composition using consumer-grade images. It introduces a dual DL pipeline with ResNet50-based single-grain classification and U-Net-based bulk segmentation to support fine-grained and aggregate quality analysis. Results show 55% overall accuracy for grain-level classification and IoU of 93% for segmentation, enabling estimation of mixture proportions in bulk samples. The approach demonstrates the practicality of accessible smartphone imagery for rice quality assessment and offers directions for improving accuracy and extending to other foods.

Abstract

This study introduces an innovative approach to classifying various types of Persian rice using image-based deep learning techniques, highlighting the practical application of everyday technology in food categorization. Recognizing the diversity of Persian rice and its culinary significance, we leveraged the capabilities of convolutional neural networks (CNNs), specifically by fine-tuning a ResNet model for accurate identification of different rice varieties and employing a U-Net architecture for precise segmentation of rice grains in bulk images. This dual-methodology framework allows for both individual grain classification and comprehensive analysis of bulk rice samples, addressing two crucial aspects of rice quality assessment. Utilizing images captured with consumer-grade cell phones reflects a realistic scenario in which individuals can leverage this technology for assistance with grocery shopping and meal preparation. The dataset, comprising various rice types photographed under natural conditions without professional lighting or equipment, presents a challenging yet practical classification problem. Our findings demonstrate the feasibility of using non-professional images for food classification and the potential of deep learning models, like ResNet and U-Net, to adapt to the nuances of everyday objects and textures. This study contributes to the field by providing insights into the applicability of image-based deep learning in daily life, specifically for enhancing consumer experiences and knowledge in food selection. Furthermore, it opens avenues for extending this approach to other food categories and practical applications, emphasizing the role of accessible technology in bridging the gap between sophisticated computational methods and everyday tasks.
Paper Structure (9 sections, 4 figures, 2 tables)

This paper contains 9 sections, 4 figures, 2 tables.

Table of Contents

  1. INTRODUCTION
  2. PROBLEM DEFINITION
  3. METHODOLOGY
  4. EXPERIMENTS
  5. Experimental Settings
  6. Datasets
  7. Evaluation Metrics
  8. Results
  9. CONCLUSION AND FUTURE WORK

Figures (4)

  • Figure 1: Schematic representation of the dual-model framework employing ResNet for single grain classification and U-Net for segmentation in bulk prediction tasks. This diagram illustrates the process flow from data collection and preprocessing through to model training and evaluation.
  • Figure 2: Visual representation of the seven Persian rice varieties classified in this study, showcasing the distinctive appearance of each variety. These images serve as a basis for both single grain classification and bulk rice prediction tasks, demonstrating the diversity and complexity of rice types found in Persian cuisine.
  • Figure 3: Comparison of a bulk rice sample with its corresponding segmentation mask. The left image showcases a diverse mixture of rice grains, while the right image illustrates the sample segmentation mask utilized and generated by the U-Net model, highlighting the model's ability to identify and segment individual grains within the bulk sample. This process is critical for estimating the percentage composition of different rice varieties in mixed samples.
  • Figure 4: Comparison of a sample bulk image with its segmentation result. The segmentation demonstrates the model's ability to accurately identify and delineate individual rice grains, facilitating precise variety composition analysis.