Tactile identification of fruits using machine learning/deep learning

• Main Author: Kua, Jia Kang
• Other Authors: Leong Wei Lin
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Electrical and electronic engineering::Computer hardware, software and systems
• Online Access: https://hdl.handle.net/10356/167727
• Institution: Nanyang Technological University

Object detection is becoming increasingly common due to the rise of automation. More autonomous robots are using object detection techniques to generate an appropriate output depending on their tasks. To perceive their surroundings and collect data, most autonomous robots use computer vision. However, there are certain situations where computer vision may not be the most optimal solution. For example, in situations where haptic properties such as the height or width of an object need to be measured, computer vision alone may not be sufficient. In such cases, other methods must be used to obtain the necessary information. Tactile identification can be used as an alternative method for autonomous robots to perceive their surroundings. This approach has been extensively researched over the years and involves using sensors to collect data, which is then trained with a machine learning or deep learning algorithm. Therefore, this project will use an in-house developed sensor together with machine learning and deep learning algorithms to classify the fruits. The data will be collected from five different types of fruit from the sensors. This data will then be used to train the machine learning and deep learning algorithms. Finally, a new batch of fruits data will be collected and tested to determine if the algorithms have successfully learned the features of each fruit. This project report will document the process of assembling the sensor, data collection, data processing and analysis of the results.