Sistem Akuisisi Gambar Menggunakan Kamera 3D Kinect V2 Untuk Memonitoring Ketinggian Tanaman Pakcoy (Brassica Rapa L.)

Azizah Alma, Ichwana Ichwana, Indera Sakti Nasution

Abstract


Abstrak. Tinggi tanaman merupakan  salah satu parameter utama pertumbuhan tanaman pakcoy yang dibudidaya secara hidroponik. Oleh karena itu perlu dilakukan pengukuran tinggi tanaman dimana membutuhkan waktu relatif lama jika di ukur secara manual. Maka solusi dari permasalahan ini adalah pengukuran otomatis menggunakan pengolahan citra utuk menggantikan tenaga manusia. Tujuan dari penelitian ini adalah merancang dan memonitoring ketinggian tanaman pakcoy sistem hidroponik menggunakan teknologi pengolahan citra dan kamera 3D Kinect v2 terintegrasi dengan komputer. Penelitian ini terdapat dua pengukuran yaitu dengan cara manual menggunakan penggaris, dan pengambilan citra menggunakan kinect v2 dengan resolusi kamera 1920 x 1080 piksel, Sampel tanaman pakcoy yang digunakan pada penelitian ini pada umur 2,3 dan 4 minggu sebanyak 60 sampel. Pengambilan citra dilakukan pada sore hari yaitu dimulai pada jam 15.00-17.00. Data citra depth yang dihasilkan diolah menggunakan software Halcon HDevelop 20.05 Student Edition dengan memasukkan algoritma dan diidentifikasi menggunakan metode Distance Transform dan Watersheds Segmentation berfungsi untuk memisahkan antara objek dengan latar belakang (background) dan membuat batas antara satu objek dengan objek lainnya yang bersentuhan ataupun berhimpit, untuk menentukan jarak tiap-tiap daun yang terdeteksi dengan kamera kinect v2 digunakan metode histogram pada pengolahan citra. Berdasarkan hasil pengujian menggunakan teknologi pengolahan citra digital sensor kinect v2 telah terbukti mampu untuk mengumpulkan data tinggi tanaman pakcoy. Persentase akurasi pada pengukuran tinggi tanaman secara actual dan kinect v2 dengan metode Root mean squre error (RMSE) yaitu <2 dengan kriteria baik, sedangkan menggunakan metode mean absolute percentage error (MAPE) yaitu sebesar 93%.

Image Acquisition System Using Kinect V2 3D Camera for Monitoring Pakcoy (Brassica Rapa L.) Plant Height

Abstract. Plant height is one of the main parameters for the growth of pakcoy plants that are cultivated hydroponically. Therefore it is necessary to measure the height of the plant which takes a relatively long time if it is actually measured. So the solution to this problem is automatic measurement using image processing to replace human labor. The purpose of this research is to design and monitor the height of the hydroponic system pakcoy plant using image processing technology and a 3D Kinect v2 camera integrated with a computer. In this study, there are two measurements, namely manually using a ruler, and image taking using kinect v2 with a camera resolution of 1920 x 1080 pixels. Pakcoy plant samples used in this study at the age of 2,3 and 4 weeks were 60 samples. Image retrieval is carried out in the afternoon, starting at 15.00-17.00. The resulting depth image data is processed using Halcon HDevelop 20.05 Student Edition software by entering an algorithm and identified using the Distance Transform and Watersheds Segmentation method, which functions to separate objects from the background and create boundaries between objects that touch or coincide. To determine the distance of each leaf detected by the kinect v2 camera, the histogram method was used in image processing. Based on the test results using digital image processing technology, the kinect v2 sensor has been proven to be able to collect data on the height of pakcoy plants. The percentage of accuracy in the actual measurement of plant height and kinect v2 using the Root mean squre error (RMSE) method is <2 with good criteria, while using the mean absolute percentage error (MAPE) method, which is 93%.


Keywords


Halcon HDevelop ; Image Processing; Kinect Camera; Pakcoy (Brassica rapa L)

Full Text:

PDF

References


Aksoy, E. E., Abramov, A., Wörgötter, F., Scharr, H., Fischbach, A., dan Dellen, B. 2015. Modeling Leaf Growth Of Rosette Plants Using Infrared Stereo Image Sequences. Computers And Electronics In Agriculture,110,78–90. Https://Doi.Org/10.1016/J.Compag.2014.10.020

Apelt, F., Breuer, D., Nikoloski, Z., Stitt, M., & Kragler, F. 2015. Phytotyping4D: A Light-Field Imaging System For Non-Invasive And Accurate Monitoring Of Spatio-Temporal Plant Growth. Plant Journal, 82(4), 693–706. Https://Doi.Org/10.1111/Tpj.12833

Chang, P., Wang, Y., & Liu, C. (2007). The development of a weighted evolving fuzzy neural network for PCB sales forecasting. 32, 86–96. https://doi.org/10.1016/j.eswa.2005.11.021

Glasbey, C. A., Polder, G., Van Der Heijden, G. W. A. M., and Song, Y. 2014. Non-Destructive Automatic Leaf Area Measurements By Combining Stereo And Time-Of-Flight Images. IET Computer Vision, 8(5), 391–403. Https://Doi.Org/10.1049/Iet-Cvi.2013.0056

Mathers, A. W., Hepworth, C., Baillie, A. L., Sloan, J., Jones, H., Lundgren, M., Sturrock, C. J. 2018. Investigating The Microstructure Of Plant Leaves In 3D With Lab-Based X-Ray Computed Tomography. Plant Methods, 14(1), 99. Https://Doi.Org/10.1186/S13007-018-0367-7.

Nguyen, T. T., Slaughter, D. C., Townsley, B., Carriedo, L., N., J. N., and Sinha, N. 2016. Comparison Of Structure-From-Motion And Stereo Vision Techniques For Full In-Field 3D Reconstruction And Phenotyping Of Plants: An Investigation In Sunflower. In 2016 ASABE International Meeting (P. 1). American Society Of Agricultural And Biological Engineers. Https://Doi.Org/10.13031/Aim.20162444593.




DOI: https://doi.org/10.17969/jimfp.v6i3.17540

Refbacks

  • There are currently no refbacks.


 


JIM Agribisnis|JIM Agroteknologi|JIM Peternakan|JIM Teknologi Hasil Pertanian|JIM Teknik Pertanian|
JIM Ilmu Tanah|JIM Proteksi Tanaman|JIM Kehutanan


E-ISSN: 2614-6053 2615-2878 Statistic Indexing | Citation


Alamat Tim Redaksi:
Fakultas Pertanian,Universitas Syiah Kuala
Jl. Tgk. Hasan Krueng Kalee No. 3, Kopelma Darussalam,
Banda Aceh, 23111, Indonesia.
Email:jimfp@usk.ac.id