International Research Journal of Engineering and Technology (IRJET) Volume: 09 Issue: 05 | May 2022
www.irjet.net
e-ISSN: 2395-0056 p-ISSN: 2395-0072
AI BASED CROP IDENTIFICATION WEBAPP Aditya Sawate, Archie Agrawal, Aanchal Saboo Under Graduate Student, Department of Computer Science & Engineering, Sipna College of Engineering & Technology, Amravati Assistant Professor A. V. Pande, Department of Computer Science & Engineering, Sipna College of Engineering & Technology, Amravati, Maharashtra, India ---------------------------------------------------------------------***--------------------------------------------------------------------algorithms like CNN the usage of the dataset containing five ABSTRACT - In general, agriculture is the backbone of India and also plays an important role in the Indian economy by providing a certain percentage of domestic products to ensure food security. But nowadays, food production and prediction are getting depleted due to unnatural climatic changes, which will adversely affect the economy of farmers by getting a poor yield and also help the farmers to remain less familiar in forecasting the future crops. This research work facilitates newbie farmers in this type of manner as to manual in sowing affordable crop through deploying machine learning, one of the superior technology in crop prediction. Convolutional Neural Network is a supervised learning algorithm that puts forth the way to achieve it . The picture records of the plants are gathered here, with an appropriate parameters like size, shape, color, and moisture content, which enables the plants to acquire a a success identity. In addition to the software, a cellular internet utility for Android is being developed. The users are encouraged to just click an image of a farm yield once it is uploaded will be taken automatically in this application to start the prediction process. Key Words: Agriculture, Crop, Prediction Algorithm, Machine Learning, Convolutional Neural Network, Mobile.
1. INTRODUCTION Machine learning is a valuable decision-making tool for predicting the type of crops and agricultural yields. To aid crop prediction studies, several machine learning methods have been used. Machine learning strategies are applied in numerous sectors, from comparing client behavior. For some years, agriculture has been using machine learning techniques. Crop prediction is certainly considered one among agriculture's complicated challenges, and numerous fashions were evolved and verified so far. Because crop manufacturing is suffering from many elements together with atmospheric conditions, kind of fertilizer, soil, and seed, this undertaking necessitates the usage of numerous datasets. This means that predicting agricultural productiveness isn't a easy process; rather, it includes a chain of complex procedures. Crop yield prediction strategies can now moderately approximate the real yield, despite the fact that greater exquisite yield prediction overall performance continues to be desired. The challenge targets to apply supervised gaining knowledge of © 2022, IRJET
|
Impact Factor value: 7.529
|
forms of crops. The outcomes monitor that the advised system gaining knowledge of technique's effectiveness is as compared to the great accuracy with precision.
2. LITERATURE REVIEW Jing Wei Tan and Siow-Wee Chang [1] suggest research on CNN is applied to extract the features from leaf images of selected tree species. Three different CNN models were used, namely, the pre-trained AlexNet CNN model, finetuned pre-trained AlexNet CNN model, and the proposed DLeaf CNN model. The extracted features were then fed into a few classification approaches for learning and training purposes. Five classifiers were employed in this research which are CNN, Support Vector Machine (SVM), Artificial Neural Network (ANN), nearest Neighbour (k-NN), and Naïve Bayes (NB). A conventional method, which segmented the leaf veins by using the Sobel edge detection technique and performed vein morphological measurements, was used for benchmarking. Based on the literature review, this is one of the first few studies, which have applied CNN in tropical tree species classification, by using both leaves morphometric and venation pattern approaches. Pankaja K and Dr. Thippeswamy G[2] suggest that endless plant species are accessible and all-inclusive. To oversee gigantic substance, the improvement of quick and successful classification techniques has transformed into a domain of dynamic exploration. As trees and plants are critical to the environment, precise Identification and grouping get important. Order strategy is helped out through several sub techniques. A recognizable proof or Classification issue is overseen by planning info information with one of the one-of-a-kind classes. In this technique, from the start, a database of leaf pictures is made, that involves pictures of test leaves with their equal plant data. Fundamental highlights are removed utilizing picture preparing methods. The highlights must be steady to make the recognizable proof framework powerful. Consequently, the plant/leaf is perceived utilizing AI procedures. In this paper, a review is introduced on the different kinds of leaf distinguishing proof procedure Thi Thanh-Nhan Nguyen Et Al [3] is a mix of profound learning and hand-planned element for plant ISO 9001:2008 Certified Journal
|
Page 1095