Classification of haploid and diploid maize seeds by using image processing techniques and support vector machines

摘要: In vivo maternal haploid technique is now widely used in advanced maize breeding programs. This shortens the period and increases efficiency of breeding. One important processes this selection seeds. The fact that performed manually reduces success causes time labor loss. For reason, it a need to develop automatic methods will save increase success. study, method was proposed classify diploid seeds by using image processing techniques support vector machines. Firstly, each seed segmented from its original image. Secondly, five different features were extracted for seed. Finally, obtained classified performance tested 10-fold cross-validation method. As result test, rate classification calculated as 94.25% 77.91%.