GÖRÜNTÜ İŞLEME TEKNİKLERİ KULLANILARAK NAPOLYON TİPİ KİRAZLARIN SINIFLANDIRILMASI

GORUNTU IŞLEME TEKNIKLERI KULLANILARAK NAPOLYON TIPI KIRAZLARIN SINIFLANDIRILMASI Ozet Insanin gorme sistemini temel alan bilgisayar destekli goruntu isleme (GI) gunluk hayatimizin her alanina girmis durumdadir. Girdi goruntulerinin degisik tekniklerle alinmasi ve islenmesi temeline dayanan goruntu isleme sonucunda, ya yeni bir goruntu elde edilir ya da goruntuden anlamli bir sonuc cikarilir. Bu tur uygulamalar ile hayatimiz olumlu etkilenmekte, islemler hem hizli hem kolay hem de ekonomik hale gelebilmektedir. Bu yuzden GI teknikleri her gecen gun gelismekte ve kendisine yeni uygulama sahalari bulmaktadir. Endustride, tarimda, muhendislikte ve bircok alanda GI calismalari yaygin olarak kullanilmaktadir. Ozellikle son yillarda tarim alanindaki goruntu isleme calismalari artan bir seviyede yayginlasmaktadir. Bu, GI’nin uygulandigi alanlara hem ekonomi hem de verim olarak olumlu yansimaktadir. Bu calismada, Turkiye’de yetistirilen Napolyon kirazlarinin boyutlarinin belirlenmesine donuk bir calisma gerceklestirilmistir. Oncelikli olarak, kirazlarin fotograflari cekilerek goruntu Matlab ortamina aktarilmistir. Alinan bu goruntudeki her bir kirazin kac kalibrede oldugu, yani capinin kac milimetre oldugu GI kullanilarak belirlenmeye calisilmistir. GI tekniklerinin uygulamaya yonelik bir calismasi ile urunlerin daha dogru siniflandirilmasi amaclanmistir. Bu sekilde, satislar sirasinda urunler hak ettikleri sekilde fiyatlandirilarak ureticinin maddi kayiplarinin onune gecilmesi saglanabilecektir. Anahtar Kelimeler: Goruntu Isleme, Kiraz, Kalibre Tespiti, Siniflandirma CLASSIFICATION FOR NAPOLEON TYPE CHERRIES BY USING IMAGE PROCESSING TECHNIQUES Abstract Computer assisted image processing (IP) based on human vision system has entered every aspect of daily life.  As a result of image processing that is based on getting and processing input images with different techniques, either a new image is received or a significant conclusion is drawn.  Our lives are affected positively via these kinds of applications and procedures can become easy and economical. Therefore, IP techniques develop day by day and find new application fields. IP works are carried out prevalently in many areas like industry, education and agriculture. Especially in recent years’ image processing works become widespread increasingly in agricultural area. This affects the areas where IP techniques are applied positively in terms of economy and productivity.  In this work, a study was performed about the determination of the size of bing cherries grown in Turkey.  Primarily the photos of the cherries were taken and the images were sent to Matlab environment.  It was tried to determine the calibre of each cherry in this image using IP techniques; that’s to say, how many millimeters of its diameter was.  It was aimed to categorise the products better by means of performance-oriented work of IP techniques.  In this way, the financial loss of the producers will be prevented by pricing the products as necessary during the sales. Keywords: Image Processing, Cherry, Calibrate Detection, Classification