FRUIT RECOGNITION APPLICATION USING MACHINE LEARNING

LOH, WENG KEONG (2023) FRUIT RECOGNITION APPLICATION USING MACHINE LEARNING. [Final Year Project Report] (Unpublished)

	PDF LOH WENG KEONG (24 pgs).pdf Download (420kB)
	PDF (Please get the password by email to repository@unimas.my , or call ext: 082-583914/3973/3933) LOH WENG KEONG ft.pdf Restricted to Registered users only Download (1MB)

Abstract

In this study, we present a fruit recognition system using machine learning techniques. The system consists of several distinct phases, including image acquisition, image pre-processing, feature extraction, classification, and results and evaluation. In the image acquisition phase, the system gathers a dataset of images that will be used for training and testing. These images may be acquired from a variety of sources, such as digital cameras or online databases. In the image pre-processing phase, the system prepares the images for further analysis by cleaning them up and standardizing them. This may include tasks such as removing noise, correcting for distortion, or resizing the images. In the feature extraction phase, the system extracts meaningful features from the images that can be used to distinguish one fruit from another. These features include color and shape features. In the classification phase, the system uses the extracted features to classify the fruits using a machine learning algorithm – a backpropagation neural network (BPNN). Finally, in the results and evaluation phase, the system assesses the accuracy and effectiveness of the classification process. Next, a fruit recognition prototype is developed to implement the trained model. The prototype is ready to use to recognize input image and display the result. Overall, our approach demonstrates a thorough and systematic approach to developing a fruit recognition system using machine learning techniques

Actions (For repository members only: login required)

View Item