Pendugaan Kualitas Fisik Biji Jagung untuk Bahan Pakan Menggunakan Jaringan Syaraf Tiruan Berdasarkan Data Citra Digital

Adrizal Adrizal, D. Anggraini, N. Novita, Santosa Santosa, Andasuryani Andasuryani


The Research intended to study the method of prediction of physical quality of corn kernel of feed stuff using Artificial Neural Network (ANN) based on variables of image data. The image data was used as input of ANN, and the character of corn kernel was the output. The variable of image were index red index, green index, blue index, hue, saturation, intensity, entropy, energy, contras, and homogeneity. The characteristic of corn were intact kernel, broken kernel, damage kernel and moldy kernel. There are two phase of application of ANN; training and validation. The training intend to calibrate the relationship between the image variable and corn kernel characteristics. The validation intend to examine the accuration of prediction. The result of research indicated that the intact kernel less accurate (70%) be predicted by image data, whereas broken kernel, damage kernel and moldy kernel can be predicted accurately (100%). The average of accuracy was 92.5 %. It was conclude that it was need to be improved the quality of image before processing the data to be input to the ANN.


artificial neural network; image processing; corn kernel; feed



  • There are currently no refbacks.


View My Stats

 Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Jurnal Peternakan Indonesia (JPI) 

Fakultas Peternakan Universitas Andalas.
Kampus Unand Limau Manis Padang-25163, Sumatera Barat, Indonesia.