International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 11 Issue: 02 | Feb 2024
p-ISSN: 2395-0072
www.irjet.net
Raisin Classification Using Machine Learning Techniques Nisha Dhital1 1Agricultural Engineer
Tribhuvan University, Institute of Engineering, Dharan ---------------------------------------------------------------------***--------------------------------------------------------------------1.1 Need For Classification
Abstract - Raisins are valuable and desired food product.
The advancement in the field of artificial intelligence has modernized classification process significantly. Raisin classification is important as precise sorting is necessary to maintain quality and for trade purposes. Traditional way of classification using manpower is time consuming and tiring. In this study, machine learning techniques like LR, KNN, DT, RF, SVM and MLP were employed on raisin data consisting seven morphological features of 900 raisin sample, to distinguish two varieties of raisin; Besni and Kecimen. After preprocessing, a cross validation of 10 fold with 80%/20% training and testing split was used to ensure generalization. The classification achieved accuracy of 87.22% with LR, 83.89% with KNN, 82.22% with DT, 86.67 %with RF, SVM, and MLP with the highest being LR with accuracy of 87.22%. Performance of these classifiers underscores the success of study.
The development of automatic raisin sorting system using machine vision is essential to address the drawbacks of manual evaluation, such as high costs, drudgery and reliability issues. This technology can enhance product quality, eliminate inconsistency and reduce dependence on labour. Classification of agricultural product is crucial for trade and marketability. It helps to maintain quality, plan logistics, plan resource allocation, set fair market value, meet food industry quality standards, and meet consumer preferences. To obtain high-quality end products, agricultural produces must be separated from the substandard ones at the initial stages. Sorting and grading are done to enhance the uniformity and commercial value of the products (7).
1.2 Artificial Intelligence (AI) in Classification Machine learning in agriculture has progressed dramatically over the past two decades, from laboratory curiosity to a practical technology in widespread commercial use. It can be far easier to train a system by showing it examples of desired input-output behavior than to program it manually by anticipating the desired response for all possible inputs (8).
Key Words: Raisin, Classification Techniques, Machine Learning, Artificial Intelligence, Evaluation Metrics
1. INTRODUCTION Raisins are dried grapes mostly obtained from different cultivars of Vitis vinifera L. and are extensively consumed worldwide (1).A portion of 100 g of raisins has 299 kcal energy, 3.3 g protein, 0.25 g total lipid (fat), 79.3 g carbohydrate, 4.5 g total dietary fiber, 65.2 g total sugars, 62 mg calcium, 1.79 mg iron, 36 mg magnesium, 98 mg phosphorus and 744 mg potassium (2).Because of its low cost and high satiety value, raisin plays a crucial role in human diets around the world (3).
The agricultural system must become more productive in output, efficient in operation, and sustainable for future generations. Artificial intelligence and machine vision are playing a key role in the world of food safety and quality assurance. AI makes it possible for computers to learn from experience, and perform most human tasks with an enhanced degree of precision and efficiency. It offers sweeping transformation with advanced approaches that will redefine the traditional pattern and limits of agriculture (9).
In 2022/2023 the total raisin production worldwide was around 1.31 million metric tons (4). The expected raisin production from Turkey and USA for year 2023/2024 is 206,300 MT and 153,000 MT respectively (5). Raisins market size was valued at USD 2.2 Billion in 2022. The raisins market industry is projected to grow from USD 2.3 Billion in 2023 to USD 3.4 Billion by 2032, exhibiting a compound annual growth rate (CAGR) of 4.81% during the forecast period (2023-2032) (6).
© 2024, IRJET
|
Impact Factor value: 8.226
Traditional methods for raisin grain classification are labor-intensive and prone to errors. Therefore artificial intelligence tools are desired in agricultural industry to develop efficient and automated techniques that can maintain product quality and align well with industry requirements. By using cameras, sensor and image processing, different features like size, morphology and color can be determined. Advanced algorithms applied to
|
ISO 9001:2008 Certified Journal
|
Page 834