Identifikasi dan Klasifikasi Benih Padi Menggunakan Analisis Citra Digital Berdasarkan Bentuk Fisik
Published: 29 May 2020
Jurnal Penelitian Pertanian Tanaman Pangan , Volume 4, pp 27-34; doi:10.21082/jpptp.v4n1.2020.p27-34
Abstract: Seed quality testing is one of aspect control and quality assurance for certified seeds. The grain shape is one of the important parameter components tested in the seed certification process especially in the field. The common method used were using human/ analysts visual observation. It has a high degree of subjectivity and low efficiency. Observation of complex samples in the field requires an alternative observation that is more subjective and accurate. An alternative technology for identifying seeds during certification and production is identification based on digital images. The purpose of this study were to identify and classify rice seeds based on physical form using digital image analysis. A total of 20 varieties with various shapes have been taken with a microscope that connected to the camera and computer. The resulting image file was analyzed using imageJ 1.51k software and analyzed statistic to discriminate the seeds tested according to their group. Results from this study indicated that digital image analysis is able to identify and classify seeds. The grouping of seeds into long, medium and round seed categories based on perimeter, circularity, AR, and round parameters with successive correlation levels is 95.4%; 82.5%; 45.3% and 38.9%. This method is more sensitive to identifying seed characteristics than eye visualization of seeds whose physical size is outside the range of numbers specified in the description. For example Cisadane, Gilirang and Ketonggo seeds.
Keywords: visualization / Digital image / method / parameter / Identify and Classify / Identifying Seeds / Analisis Citra / seeds tested
Scifeed alert for new publicationsNever miss any articles matching your research from any publisher
- Get alerts for new papers matching your research
- Find out the new papers from selected authors
- Updated daily for 49'000+ journals and 6000+ publishers
- Define your Scifeed now
Click here to see the statistics on "Jurnal Penelitian Pertanian Tanaman Pangan" .