Application of image analysis for maturity and chilling injury classification of 'Phulae' pineapple

Name: Ullah, Habib

Address: 333 Moo1, Thasud, Muang, Chiang Rai 57100

Organization : Mae Fah Luang University. School of Agro-Industry. Technology Management of Agricultural Produces

Email : ask.library@mfu.ac.th

keyword: Phulae pineapple

LCSH: Pineapple -- Postharvest technology

Classification :.LCCS: SB375

LCSH: Pineapple -- Quality

LCSH: Pineapple -- Biotechnology

Abstract: In this research, image analysis technique was developed as an alternative tool of visual maturity classification and chilling injury (CI) evaluation of ‘Phulae’ pineapple harvested in summer (5th June 2017), rainy (19th September 2017), and winter (16th January 2018) season. Generally, pineapple maturity classification is done manually by experienced persons based on fruit shell color which is not a precise technique. Sometimes, the shell color appears green with inside yellow and sweet flesh concerned as a marketing problem in pineapple industry. Image analysis pineapple maturity classification and chilling injury evaluation is more effective, rapid, and consistent technique in industries use than human judgment due to factors involved such as fatigue and lack of attention. In addition, artificial neural network (ANN) modelling was developed for ‘Phulae’ pineapple fruit maturity classification and CI evaluation. ‘Phulae’ pineapple harvested at three different maturity stages based on percentage yellow shell color i.e., green (1 to 10%), green-yellow (11 to 75%) and yellow (>75%) were investigated by using image analysis technique. The percentage of yellow shell area and the extracted RGB color values were used for pineapple maturity classification. The percentage of yellow shell area significantly increased (P<0.05) as fruit aged. Moreover, the R (red) value seemed to be the best parameter for distinguish pineapple maturity comparing to those of G and B values. In terms of color conversion, the range of correlation coefficient (r) values between colorimeter CIELAB and converted CIELAB values showed highly correlation ranging from -0.704 to 0.972. In terms of CI determination, the fruit were stored at 10 ºC for four weeks and CI was evaluated after keeping pineapples at ambient temperature for 3 days. The percentage of CI area and fractal dimension (FD) to indicate CI severity were weekly evaluated. The percentage of CI area incidence ranged 0.00-71.00%. On the other hand, the FD values increased with the increasing of CI severity which ranged 0.00-1.91. Moreover, the accuracy of the CI severity level evaluation was 86.11% and 66.07% for the percentage CI area and FD determinations, respectively, as compared to visual scoring. Ninety datasets of the experimental data were used for developing ANN model to evaluate maturity stages of ‘Phulae’ pineapple. The 6 – nodes – one hidden layer architecture was the best architecture that provided the lowest root mean square error (RMSE) (0.26 for maturity stage classification) and highest coefficient of determination (R2) value (0.89 for maturity stage classification). In terms of CI evaluation using ANN model, the 20 – nodes – one hidden layer was the most suitable architecture with the lowest RMSE of 0.39 and the highest R2 of 0.89. The proposed methods showed the promising efficacy to distinguish maturity and would be possibly applied for the commercial used.

Mae Fah Luang University. The Learning Resources and Education Media Center

Address: Chiang Rai

Email: library@mfu.ac.th

Name: Rattapon Saengrayap

Role: Advisor

Email : ask.library@mfu.ac.th

Name: Saowapa Chaiwong

Role: Co-Advisor

Email : ask.library@mfu.ac.th

Created: 2018

Modified: 2019-05-22

Issued: 2019

Issued: 2019-05-22

วิทยานิพนธ์/Thesis

application/pdf

CallNumber: Thesis SB375 U410a 2018

eng

DegreeName: Master of Science

Level: Master's Degree

Descipline: Technology Management of Agricultural Produces

©copyrights Mae Fah Luang University

ลำดับที่.	ชื่อแฟ้มข้อมูล	ขนาดแฟ้มข้อมูล	จำนวนเข้าถึง	วัน-เวลาเข้าถึงล่าสุด
1	127031.pdf	8.03 MB	3	2024-03-30 18:47:50

ใช้เวลา

0.039646 วินาที