Visual Dataset for Spice Recognition: Enhancing Classification of Common Spices

This research on "Spice Dataset Classification" investigates machine learning methods for accurately classifying spices from a dataset of 4,134 images. The dataset includes images of Black Cardamom, Black Pepper, Cinnamomum Tamala, Cinnamon, Clove, Fenugreek, Garlic, and Red Chili. This study aims to advance automated spice identification, benefiting food quality control and culinary science. The distribution of spice samples is as follows:

Black Cardamom: 424 samples
Black Pepper: 1100 samples
Cinnamomum Tamala: 314 samples
Cinnamon: 301 samples
Clove: 958 samples
Fenugreek: 414 samples
Garlic: 309 samples
Red Chili: 314 samples

Purpose:
The purpose of this research is to develop a reliable and accurate machine learning model for the classification of spices based on image data. By automating the recognition process, this study aims to facilitate the identification of various spices—such as Black Cardamom, Black Pepper, Cinnamomum Tamala, Cinnamon, Clove, Fenugreek, Garlic, and Red Chili thereby supporting applications in food quality control, inventory management, and culinary research. This work seeks to enhance efficiency in spice identification, reduce human error in classification tasks, and contribute to advancements in food science and automated recognition systems.

本研究针对“香料数据集分类”展开，旨在探究从包含4,134张图像的数据集中准确分类香料的方法。该数据集涵盖了黑姜、黑胡椒、丁子香、肉桂、丁香、芝麻菜、大蒜和红辣椒等香料的图像。本研究旨在推进香料自动识别技术，从而促进食品质量控制与烹饪科学的发展。香料样本的分布情况如下：
黑姜：424个样本
黑胡椒：1100个样本
丁子香：314个样本
肉桂：301个样本
丁香：958个样本
芝麻菜：414个样本
大蒜：309个样本
红辣椒：314个样本
研究目的：
本研究的目的是开发一种基于图像数据的香料分类的可靠且精确的机器学习模型。通过自动化识别过程，本研究旨在促进各种香料（如黑姜、黑胡椒、丁子香、肉桂、丁香、芝麻菜、大蒜和红辣椒等）的识别，从而支持食品质量控制、库存管理和烹饪研究等应用。本研究旨在提高香料识别的效率，减少分类任务中的人为错误，并为食品科学和自动化识别系统的进步做出贡献。