SETIAWAN, Aji and Adi, Kusworo and Widodo, Catur Edi (2024) DETEKSI OBJEK ASING PADA BERAS MENGGUNAKAN PENGOLAHAN CITRA DAN CONVOLUTIONAL NEURAL NETWORK (CNN). Doctoral thesis, UNIVERSITAS DIPONEGORO.

	Text COVER dll.pdf Download (1MB)
	Text 6. BAB 1.pdf Download (457kB)
	Text 7. BAB 2.pdf Download (2MB)
	Text 8. BAB 3.pdf Restricted to Repository staff only Download (1MB)
	Text 9. BAB 4.pdf Restricted to Repository staff only Download (3MB)
	Text 10. BAB 5.pdf Restricted to Repository staff only Download (229kB)
	Text 11. Daftar Pustaka.pdf Download (461kB)
	Text 12. Lampiran.pdf Restricted to Repository staff only Download (849kB)

Abstract

Beras merupakan salah satu komponen bahan pokok yang termasuk dalam isu ketahanan pangan dunia dalam program sustainable development goals (SDGs). Tingginya permintaan kebutuhan beras sejalan dengan tingginya kualitas beras yang aman dari benda asing yang dapat membahayakan konsumen. Dampak buruk yang timbul dari buruknya kualitas beras dengan adanya beberapa beras cacat dan benda asing yang masuk mengakibatkan ketidakpercayaan konsumen terhadap produk yang dihasilkan karena membahayakan kesehatan konsumen dan menimbulkan kerugian ekonomi bagi petani. Penelitian yang dilakukan membahas tentang upaya dalam mendeteksi objek asing pada kumpulan beras dengan beberapa parameter benda asing beras cacat, batu dan gabah dengan menggunakan pendekatan ekstraksi ciri pengolahan citra dan Convolutional Neural Network (CNN). Hasil penelitian menunjukan model yang dibangun mendapatkan nilai akurasi 97%.
Kata kunci: Beras, Benda Asing, Pengolahan Citra, Convolutional Neural Network (CNN).

Rice is one of the staple components included in the issue of world food security in the sustainable development goals (SDGs) program. The high demand for rice is in line with the high quality of rice which is safe from foreign objects that can harm consumers. The negative impact that arises from the poor quality of rice with some defective rice and foreign objects entering it results in consumer distrust of the product produced because it endangers consumer health and causes economic osses for farmers. The research carried out discussed efforts to detect foreign objects in a collection of rice with several foreign object parameters of defective rice, stones and grain using an image processing feature extraction approach and a Convolutional Neural Network (CNN). The research results showed that the model built had an accuracy value of 97%.
Keyword: Rice, Foreign Objects, Image Processing, Convolutional Neural Network (CNN).

Actions (login required)

View Item