IDENTIFIKASI SPERMA SAPI NORMAL DAN ABNORMAL MENGGUNAKAN ALGORITMA JARINGAN SARAF TIRUAN

Orlando, Geovani (2015) IDENTIFIKASI SPERMA SAPI NORMAL DAN ABNORMAL MENGGUNAKAN ALGORITMA JARINGAN SARAF TIRUAN. S1 thesis, Universitas Sultan Ageng Tirtayasa.

Text
IDENTIFIKASI SPERMA SAPI NORMAL DAN ABNORMAL.pdf - Published Version
Restricted to Registered users only
Download (5MB)

Abstract

An analysis of normal and abnormal of spermatozoa has been done based on feature sizes that are the length and width of the head, mid-piece and the length of the tail. However, normal and abnormal spermatozoa are also depending on the form. Therefore, this study aimed to identification normal and abnormal sperm of bulls using algorithm artificial neural network based on shape features. The object of research is the image of a bull sperm obtained from the website of the University of Wisconsin-Madison department of animal sciences United States, which consists of 30 images of a bull sperm normal and 30 images of a bull sperm abnormal. Image segmentation method is to separate the sperm from the background. Getting the edge detection by using Canny edge detection method. Canny edge is an early stage in the chain coding by using chain freeman code method. Chain code is used to get the premises feature extraction Elliptical Fourier Descriptor associated with harmonic fourier coefficients to-n namely , , dan . Harmonic fourier coefficients are used for the identification of normal and abnormal sperm bull by using algorithm artificial neural network. The identification results of three tests performed showed the best accuracy on Fourier harmonic is equal to 6 (N = 6), amounting to 80% thereby shape feature can be used to identification the bull normal and abnormal sperm.

Item Type:

Thesis (S1)

Contributors: