Please use this identifier to cite or link to this item: http://repository.ipb.ac.id/handle/123456789/115234
Full metadata record
DC FieldValueLanguage
dc.contributor.advisorJaya, Indra-
dc.contributor.advisorRahmat, Ayi-
dc.contributor.authorAhmadi, Yudhi-
dc.date.accessioned2022-11-10T23:38:26Z-
dc.date.available2022-11-10T23:38:26Z-
dc.date.issued2022-
dc.identifier.urihttp://repository.ipb.ac.id/handle/123456789/115234-
dc.description.abstractPemantauan pertumbuhan kerapu secara teratur akan mendapatkan produksi ikan yang lebih efisien dan sukses dalam budidaya keramba. Penelitian ini menggunakan dua kelompok ikan kerapu hidup yang berbeda ukuran untuk mendapatkan gambaran pertumbuhannya. Tujuan penelitian ini yaitu mengaplikasikan algoritma koreksi gambar de-haze dan kecerdasan buatan (deep learning) untuk mengestimasi panjang ikan kerapu hidup di dalam KJA dari hasil perekaman kamera stereo UTS (Underwater Televisual System). Untuk memperkirakan panjang ikan kerapu hidup di dalam keramba, kami telah merekam video menggunakan kamera stereo yang merupakan bagian dari Underwater Televisual System (UTS). UTS dapat mengestimasi kedalaman jarak titik objek ke UTS dengan menggunakan dua kamera kiri dan kanan, sehingga dapat diperoleh persepsi 3D, dan penglihatan triangulasi sinar dari berbagai sudut pandang. Data pose yang dikumpulkan berjumlah 500 gambar kemudian dilakukan koreksi gambar de-haze, data preprocessing, dan diberi label. Panjang ikan kerapu diukur menggunakan algoritma SSD Mobilenet v3. Nilai training loss yang diperoleh 0.06%, accuracy 92.00%, precision 92,00%, recall 100%, dan F1-score 95.83%. Hasil pengukuran menunjukkan bahwa rata-rata panjang kedua kelompok ikan kerapu yang berbeda ukuran masing-masing adalah 12.65 ± 3.81 cm dan 26.00 ± 19.79 cm. Kesimpulannya, pengukuran panjang ikan kerapu hidup dapat dilakukan secara langsung dan akurat menggunakan kombinasi koreksi gambar de-haze dan SSD Mobilenet v3.id
dc.description.abstractRegular monitoring of the growth of the grouper will ensure the efficient and successful production of the fish in cage culture. In this study we used two different size of fish groups of free-swimming grouper to obtain its estimate length. The purpose of this study is to apply a de-haze image correction algorithm and artificial intelligence (deep learning algorithm) to estimate the length of free-swimming grouper in the KJA from the results of the UTS (Underwater Televisual System) stereo camera recording. To estimate the fish length of the free-swimming grouper in the cage, we have recorded a video using a stereo camera Underwater Televisual System (UTS). UTS can estimate the depth of a point object from the UTS using two cameras left and right, hence its 3D perception, and by triangulation of rays from multiple viewpoints. There were 500 images of pose collected and then dehaze to correct the image, preprocessed, and labeled. To measuring the length of the grouper, we used SSD Mobilenet v3 algorithm. The training loss value obtained is 0.06%, accuracy 92.00%, precision 92,00%, recall 100%, and F1-score 95.83%. The measurement results showed that the average length of the two groups different size of free-swimming grouper was 12.65 ± 3.81 cm and 26.00 ± 19.79 cm, respectively. In conclusion, the measurement of the free-swimming grouper length can be performed directly and accurately using combination of de-hazing image correction and SSD Mobilenet v3.id
dc.language.isoidid
dc.publisherIPB Universityid
dc.titleEstimasi Panjang Ikan Kerapu Hidup di Keramba Jaring Apung Menggunakan Koreksi Gambar De-Haze dan Metode Deep Learningid
dc.title.alternativeFish Length Estimation of Free-Swimming Grouper in the Cage Culture Using De-haze Image Correction, and Deep Learning Methodsid
dc.typeUndergraduate Thesisid
dc.subject.keywordgrouperid
dc.subject.keywordfree-swimming grouper lengthid
dc.subject.keywordcamera underwater televisual systemid
dc.subject.keywordSSD Mobilenet v3id
Appears in Collections:UT - Marine Science And Technology

Files in This Item:
File Description SizeFormat 
Cover, Lembar Pengesahan, Prakarta, Daftar Isi.pdf
  Restricted Access
Cover805.8 kBAdobe PDFView/Open
C54180030_Yudhi Ahmadi.pdf
  Restricted Access
Fullteks1.74 MBAdobe PDFView/Open
Lampiran.pdf
  Restricted Access
Lampiran717.03 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.