Identification of dried sea cucumber species using color and texture extraction / Novie Dave B. Perez
Material type:
Item type | Current library | Collection | Call number | Status | Date due | Barcode | |
---|---|---|---|---|---|---|---|
![]() |
Commission on Higher Education Thesis | Thesis and Dissertation | LG 996 2018 C6 P47 (Browse shelf(Opens below)) | Storage Area (Restricted Access) | CHEDFR-000316 | ||
![]() |
Commission on Higher Education Digital Thesis and Dissertation | Digital Thesis and Dissertation | LG 996 2018 C6 P47 (Browse shelf(Opens below)) | Available (Room Use Only) | DCHEDFR-000059 |
Thesis (Master of Science in Computer Engineering) -- MApua University, July 2018.
Sea cucumber is one of the highest valued marine products. The purpose of this study is to design a system that will be used to capture dried sea cucumber samples and implement color and texture feature extraction for specie identification. The system comprised of hardware and software components in which the focus of this study is to implement the algorithm of Color and Texture Extraction to be used in training and testing a Naive Bayes classifier. Haralick texture was extracted using GLCM (Gray-Level Co-occurrence Matrix) method and Color features were taken from the converted RGB to HSV color space model. In texture extraction, each of the image were converted into its gray level counterpart then 13 texture feature classes were extracted from it. In HSV color space, 512 color features were extracted from a quantized 2: )x2: pixel image. The text re and HSV color features were fed into a Naive Bayes Classifier for training and testing. Upon testing the prototype and employing 4-Fold cross validation, the result turns out to be good as it yields a 75.26% overall accuracy in identifying dried sea cucumber species.
There are no comments on this title.