PERANCANGAN SISTEM IDENTIFIKASI KEMATANGAN BUAH PADA TANAMAN HORTIKULTURA MENGGUNAKAN METODE DETEKSI WARNA

RAMADHONI, TRI SATYA and Zulkarnain, Zulkarnain and Yani, Irsyadi (2016) PERANCANGAN SISTEM IDENTIFIKASI KEMATANGAN BUAH PADA TANAMAN HORTIKULTURA MENGGUNAKAN METODE DETEKSI WARNA. Undergraduate thesis, Sriwijaya University.

[thumbnail of RAMA_21201_03121005070_0010058102_0025127104_01_front_ref.pdf]
Preview
Text
RAMA_21201_03121005070_0010058102_0025127104_01_front_ref.pdf - Accepted Version
Available under License Creative Commons Public Domain Dedication.

Download (795kB) | Preview
[thumbnail of RAMA_21201_03121005070_0010058102_0025127104_02.pdf] Text
RAMA_21201_03121005070_0010058102_0025127104_02.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (1MB) | Request a copy
[thumbnail of RAMA_21201_03121005070_0010058102_0025127104_03.pdf] Text
RAMA_21201_03121005070_0010058102_0025127104_03.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (330kB) | Request a copy
[thumbnail of RAMA_21201_03121005070_0010058102_0025127104_04.pdf] Text
RAMA_21201_03121005070_0010058102_0025127104_04.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (4MB) | Request a copy
[thumbnail of RAMA_21201_03121005070_0010058102_0025127104_05.pdf] Text
RAMA_21201_03121005070_0010058102_0025127104_05.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (51kB) | Request a copy
[thumbnail of RAMA_21201_03121005070_0010058102_0025127104_06_ref.pdf] Text
RAMA_21201_03121005070_0010058102_0025127104_06_ref.pdf - Bibliography
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (54kB) | Request a copy
[thumbnail of RAMA_21201_03121005070_0010058102_0025127104_07_lamp.pdf] Text
RAMA_21201_03121005070_0010058102_0025127104_07_lamp.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.

Download (8MB) | Request a copy

Abstract

On the handling of post harvest horticulture crops of fruit, one problem that occurs is the lack of quality in sorting between ripe fruit and not overcooked. The process of identifying fruits and vegetables are done manually is inefficient and less conscientious for a considerable sum. Based on the background that have been described previously, then the author is interested in doing more research on applicative that is by designing a system of camera-equipped sensor sorter image processing which is in the process of with computer vision in order to make it easier for humans especially farmers and middle-class society down in doing the process sort of horticultural plant types automatically bear fruits. This research standard begins with a search, learn and understand the literature study be reputed journals or scientific paper that has been there in order to get a new lessons from previous research. One of the journal of research on the improvement of the quality of cacao crops using fuzzy expert system with mamdani method by Yazdi and Handono. He said that the quality of cacao influenced by the contents of the level of water, fungi, dirt and the number of seeds in a hundred grams. The research done on 2013 aims to produce a cleverly maximizes both interior of experts to determine the quality of cacao. Then after getting the journals and papers of scientific research continued with performs a number of preparation that is gathering materials test horticultural fruits. This research uses fruit star fruit, citrus fruits and tomatoes. After that prepare the program for image retrieval by using the Matlab software R2014. The taking of the image of fruits using webcams with 16 Megapixel resolution in static condition. This research will use the 450 image of fruit. On each fruit taken 150 images. Fruit star fruit 150 image, citrus fruits and tomato image 150 150 images. With each of the 50 images of young fruit, 50 half-cooked fruit image, 50 image so it becomes ripe fruit 150 image of each fruit. Image retrieval using color detection method of RGB (red, green, blue). The next process is the creation of a database. A database is a collection of color values in the form of red, green and blue color taken from each image. The database is created in the form of tables arranged by type and category of fruit ripeness levels. After the database is arranged, the next process is the process of training. The training was conducted by searching for the limitation of the minimum value xi and the maximum value based on the level of maturity of the fruit that is the fruit of a young, half-cooked fruit and ripe fruit of every kind of fruit with a limit value of data extracted from the database. This training using Adaptive Neuro Fuzzy Inference System (ANFIS). Training is done to get a rule based fuzzy (base rules) from data that is trained. In this study formed the membership function 9, 27 and 1 rules base the value of output. Further identification of program creation by using the existing training data. After the program is created using data trained that have been prepared and testing can be performed. The tests are performed on a prototype of the belt conveyors with a webcam installed in the static condition conveyors. Webcam is used as a color detection sensor. The process of the test was conducted with how to identify the fruit based on the level of ripness repeatedly. The tests are performed as much as 135 times. Each fruit is tested as much as 15 times that is 15 times the young fruit, 15 times the half-cooked fruit and ripe fruit 15 times where every done is done with a different position. The results of this testing the program successfully identify as many as 118 times with 17 times the program one detect due to several factors such as the lighting of redundant and vice versa. On the star fruit percentage test success is 86,66%. On the citrus fruit percentage of success in detecting the fruit based on the level of ripness is 88,88%. On the fruit of the tomato percentage test success is 86,66%. And testing on the whole fruit test the percentage of success obtained is 87,40%. Based on the test results it can be concluded that the identification system can separate pieces based on the level of maturity that young fruit, halfripe fruit and ripe fruit well.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: image processing, rgb, identification, webcam, Adaptive Neuro Fuzzy Inference System (ANFIS), color detection
Subjects: T Technology > TJ Mechanical engineering and machinery > TJ1-1570 Mechanical engineering and machinery
T Technology > TJ Mechanical engineering and machinery > TJ181-210 Mechanical movements
Divisions: 03-Faculty of Engineering > 21201-Mechanical Engineering (S1)
Depositing User: Users 2681 not found.
Date Deposited: 27 Nov 2019 08:34
Last Modified: 27 Nov 2019 08:34
URI: http://repository.unsri.ac.id/id/eprint/18806

Actions (login required)

View Item View Item