Hyperspectral imaging on agrofood products

• Main Author: Loh, Nicholas Chin Kiat
• Other Authors: Murukeshan Vadakke Matham
• Format: Final Year Project
• Language: English
• Published: 2017
• Subjects: DRNTU::Engineering
• Online Access: http://hdl.handle.net/10356/72966
• Institution: Nanyang Technological University

In today society, the technology is constantly changing and improving. The world population is also increasing at an alarming rate. With a sigificant rise in the food demands, food industries are constantly coming up with technologies that improve productivity and efficiency. Thus this brings about the concern of the quality of the food produced. With a fast productive rate, there is a need for a fast and efficient food quality assessment technology. Food quality assessment is imperative to ensure that the food produce are safe for consumption. Being a fast and efficient non-destructive method, hyperspectral imaging is the ideal technique engaged for such assessment. Hyperspectral imaging is considered one of the most anticipated innovation as it has a lot of potential exceeding that of the capability of panchromatic imagers. The advantageous is mainly a result of its abilty to classify, quantify and assign chemical, physical and biological characteristics in every single individual pixel of a scanned sample. Hyperspectral imaging technique produce a massive amount of data, thus there is a need for a fast and efficient data processor software like MATLAB to be coupled along with it. However the downside of this system is that in the midst of scanning, there is bound to be problems affecting the process as a result of the rate of the information being enlarged by the number of spectral planes in the cube. It is shown that even the smallest difference in the intensity of the spectral of any given wavelengths between the same surface within the user-defined region-of-interest (ROI). In this research, a spatial scanning pushbroom hyperspectral imaging (HSI) framework which consists of a video camera and a detector camera will be utilised for data extraction. This framework is also coupled with data processing software like MATLAB for efficient data processing. This technique has the potential to extent beyond the application for agrofood quality assessment, it is also effective in the security and biomedical industries.