Show simple item record

dc.contributor.advisorSolahudin, Mohamad
dc.contributor.authorJULIANTI, SELFI DWI
dc.date.accessioned2026-07-16T23:00:54Z
dc.date.available2026-07-16T23:00:54Z
dc.date.issued2026
dc.identifier.urihttp://repository.ipb.ac.id/handle/123456789/174903
dc.description.abstractKandungan klorofil merupakan salah satu indikator penting kondisi fisiologis tanaman yang umumnya diukur menggunakan SPAD meter. Namun, metode tersebut memiliki keterbatasan karena hanya mengukur pada titik tertentu dan memerlukan kontak langsung dengan daun. Penelitian ini bertujuan merancang dan membangun sistem machine vision berbasis citra RGB dan Artificial Neural Network (ANN) untuk menduga kandungan klorofil tanaman daun dewa secara nondestruktif serta mengevaluasi pengaruh pemberian biostimulan. Penelitian dilakukan menggunakan 24 tanaman yang dibagi ke dalam tiga kelompok perlakuan, masing-masing terdiri atas delapan tanaman. Citra tanaman diakuisisi menggunakan webcam dalam capture box dengan pencahayaan terkontrol, kemudian dilakukan segmentasi berbasis HSV dan ekstraksi fitur warna. Model ANN dikembangkan menggunakan 144 data representatif hasil seleksi K-means dari total 336 data. Hasil penelitian menunjukkan bahwa sistem yang dikembangkan mampu menduga nilai SPAD dengan akurasi yang baik, ditunjukkan oleh nilai RMSE sebesar 3,07 SPAD, MAE sebesar 2,55 SPAD, dan koefisien determinasi (R²) sebesar 0,912. Selain itu, pemberian biostimulan berbasis ekstrak alga menunjukkan kecenderungan meningkatkan nilai SPAD sebesar sekitar 5–7 unit dibandingkan kontrol. Hasil ini menunjukkan bahwa sistem machine vision berbasis ANN berpotensi menjadi metode alternatif yang cepat, nondestruktif, dan ekonomis untuk pemantauan kandungan klorofil tanaman daun dewa.
dc.description.abstractChlorophyll content is an important indicator of plant physiological status and is commonly measured using a SPAD meter. However, this method is limited because it only measures specific leaf points and requires direct contact with the plant. This study aimed to design and develop an RGB image-based machine vision system integrated with an Artificial Neural Network (ANN) to estimate chlorophyll content in daun dewa (Gynura segetum) non-destructively and to evaluate the effect of biostimulant application. The study involved 24 plants divided into three treatment groups, each consisting of eight plants. Plant images were acquired using a webcam inside a controlled lighting capture box, followed by HSV-based segmentation and color feature extraction. The ANN model was developed using 144 representative data selected through K-means clustering from a total of 336 observations. The developed system successfully estimated SPAD values with good accuracy, achieving an RMSE of 3.07 SPAD, an MAE of 2.55 SPAD, and a coefficient of determination (R²) of 0.912. Furthermore, seaweed extract-based biostimulant application tended to increase SPAD values by approximately 5–7 units compared with the control group. These findings indicate that the ANN-based machine vision system has potential as a rapid, non-destructive, and cost-effective alternative for monitoring chlorophyll content in daun dewa plants.
dc.description.sponsorshipCenter for Plant Phenotyping and Controlled Environment Agriculture (CPP-CEA)
dc.language.isoid
dc.publisherIPB Universityid
dc.titleRancang Bangun Machine Vision untuk Menduga Kandungan Klorofil Tanaman Daun Dewa (Gynura segetum) pada Pemberian Biostimulanid
dc.title.alternativeDesign and Development of a Machine Vision System for Estimating Chlorophyll Content in Daun Dewa (Gynura segetum) Under Biostimulant Application
dc.typeSkripsi
dc.subject.keywordArtificial Neural Networkid
dc.subject.keywordBiostimulanid
dc.subject.keywordGynura segetumid
dc.subject.keywordmachine visionid
dc.subject.keywordSPADid
dc.subtypeUndergraduate Theses


Files in this item

Thumbnail
Thumbnail
Thumbnail

This item appears in the following Collection(s)

Show simple item record