Automated Grading of Edible Birds Nest



Loh, Selina Xin Ci (2018) Automated Grading of Edible Birds Nest. Final Year Project (Bachelor), Tunku Abdul Rahman University College.

[img] Text
Restricted to Registered users only

Download (3MB)


Edible Birds Nest (EBN) is a famous Chinese delicacy first consumed by the Chinese royalty back in the days. Till today, EBN is well known for its nutritional supplements and also the ability to aid recovery from chronic illnesses. EBNs are graded according to the quality of the EBN. The quality of the EBN is determined by a few factors such as the shape, size, colour and the cleanliness of the bird nest. The grading of EBN depends on the experience, knowledge and opinion of the human expert. Therefore, grading of EBN is very subjective. The purpose of this research is to develop an automated grading of edible bird nest system. Image processing techniques such as thresholding, Sobel edge detection, erosion and HSV colour model were used for the feature extraction. Features extracted from the EBN images are the number of impurities pixels, total area, curvature and the colour of the EBN. The four features are then used with the K-Nearest Neighbour (KNN) classifier to grade the EBNs. Each feature was tested with the KNN classifier to obtain the accuracy of the feature extracted for classification of the EBN. By varying the number of neighbours, the accuracy of the classifier was noted. The number of neighbours that achieved the highest accuracy was used to test the other parameters of the KNN classifier. Furthermore, the classifier was tested with 3 different ratios of training and testing data (60%, 70%, 80% training and 40%, 30% and 20% testing). Pre-processing techniques were applied to the feature data before testing on the KNN classifier. The parameters of the KNN classifier were adjusted to achieve the highest possible accuracy with the features. The final system is to complement the human expert whenever an additional opinion is needed when grading the EBN. It can also serve as a reference for new operators that have just completed their training on grading the EBNs.

Item Type: Final Year Project
Subjects: Technology > Mechanical engineering and machinery
Technology > Electrical engineering. Electronics engineering
Faculties: Faculty of Engineering and Technology > Bachelor of Engineering (Honours) Mechatronic
Depositing User: Library Staff
Date Deposited: 10 Oct 2018 08:05
Last Modified: 10 Oct 2018 08:05