Normal view MARC view ISBD view

Raspberry pi waste segregator using convolutional neural network algorithm / Mary El F. Bonza, Jazie T. Dante, and Anerol Leuna Alva S. Sergio

By:

Bonza, Mary El F [author.]

Contributor(s):

Material type: Text

TextDescription: 119 leavesContent type:

text

Media type:

unmediated

Carrier type:

volume

Dissertation note: Undergraduate Thesis (BSCS) - Computer Science Camarines Sur Polytechnic Colleges, 2023 Abstract: One of the most prevalent issues in schools nowadays is the enormous amount of waste generated. This issue affects waste management, which causes pollution and other infec-tious diseases that negatively affect daily life. This research was conducted to develop a prototype that would help to determine and classify waste through image processing using the CNN algorithm. The researchers used an experimental research design to perform re-search based on objectives. Also, the researchers collected a total of 5,721 image datasets in school and other images gathered from a reliable internet source. The evaluation method that is utilized in this study is Lobe, to build and train image processing models. Also, con-fusion matrix to evaluate the performance of the classification of the aforementioned algo-rithm. The Raspberry Pi camera is used to capture and detect waste accurately, the camera must be placed in a stable space. Once the waste is captured and detected the algorithm classifies if it is biodegradable, non-biodegradable, or hazardous then the LED light blinks based on the trash bin it belongs to. The model has achieved 45% accuracy. Therefore, the developed prototype successfully shows how efficient and effective the Convolutional Neural Network algorithm is used in image processing to classify waste images into three classifications. In addition, future researchers could utilize another existing algorithm in comparison with this research algorithm.

Tags from this library: No tags from this library for this title. Log in to add tags.

Holdings
Item type	Current library	Shelving location	Call number	Copy number	Status	Date due	Barcode
Unpublished Materials	Main Library	Undergraduate Thesis Section	UTH BSCS B644r 2023 (Browse shelf(Opens below))	1-1	Not for loan		006692UM

Browsing Main Library shelves, Shelving location: Undergraduate Thesis Section Close shelf browser (Hides shelf browser)

Previous	No cover image available	No cover image available	No cover image available	No cover image available	No cover image available	No cover image available	No cover image available	Next
Previous	UTH BSCS B46a 2019 ACUVJA: attendance checker using Viola-Jones algorithm	UTH BSCS B611d 2019 Decision support system in job qualification using Naive Bayes algorithm	UTH BSCS B640d 2019 Development of location recorder with augmented reality using simultaneous location and mapping	UTH BSCS B644r 2023 Raspberry pi waste segregator using convolutional neural network algorithm /	UTH BSCS B76d 2021 Development of arduino-based obstacle avoidance robot using genetic algorithm	UTH BSCS B792e 2017 eConDi: An implementation of greedy algorithm to a condiment dispenser using arduino/	UTH BSCS B863i 2023 IT support helpdesk ticket service classification using recurrent neural network /	Next

Undergraduate Thesis (BSCS) - Computer Science Camarines Sur Polytechnic Colleges, 2023

Includes bibliographic references.

One of the most prevalent issues in schools nowadays is the enormous amount of waste generated. This issue affects waste management, which causes pollution and other infec-tious diseases that negatively affect daily life. This research was conducted to develop a prototype that would help to determine and classify waste through image processing using the CNN algorithm. The researchers used an experimental research design to perform re-search based on objectives. Also, the researchers collected a total of 5,721 image datasets in school and other images gathered from a reliable internet source. The evaluation method that is utilized in this study is Lobe, to build and train image processing models. Also, con-fusion matrix to evaluate the performance of the classification of the aforementioned algo-rithm. The Raspberry Pi camera is used to capture and detect waste accurately, the camera must be placed in a stable space. Once the waste is captured and detected the algorithm classifies if it is biodegradable, non-biodegradable, or hazardous then the LED light blinks based on the trash bin it belongs to. The model has achieved 45% accuracy. Therefore, the developed prototype successfully shows how efficient and effective the Convolutional Neural Network algorithm is used in image processing to classify waste images into three classifications. In addition, future researchers could utilize another existing algorithm in comparison with this research algorithm.

There are no comments on this title.

to post a comment.