TreeLogy
2,0star
127 ta sharh
10 ming+
Yuklanmalar
Hamma uchun
info
Bu ilova haqida
“Treelogy” is a mobile application, which can perform leaf-based tree identification among tree species in Turkey using one picture of a given leaf.
There is only handful of applications interested in tree identification and they are designed primarily for detecting North American and European tree species. There is no application which has a good performance and localization support for identifying tree species native to Turkey. This project is aimed to fill this gap.
While we are constructing this application, we worked on supervised learning for classification task. We focused on both Deep Learning (specifically Deep Convolutional Neural Networks) and Support Vector Machines. Tree identification process uses leaf image features gathered from Caffe, a convolutional neural network framework, and our image processing module.
After several experiments, we reached the optimal classification accuracy of 93.59% for 57 tree species. Experiments involve 16096 training and 3020 testing leaf images. According to our findings, we come to the following conclusion. Certain image processing procedures for extracting features such as shape and texture descriptors, which we have used in our project, does not produce features as feasible as convolutional neural networks.
"Created by group paY inekereG" => METU Students
There is only handful of applications interested in tree identification and they are designed primarily for detecting North American and European tree species. There is no application which has a good performance and localization support for identifying tree species native to Turkey. This project is aimed to fill this gap.
While we are constructing this application, we worked on supervised learning for classification task. We focused on both Deep Learning (specifically Deep Convolutional Neural Networks) and Support Vector Machines. Tree identification process uses leaf image features gathered from Caffe, a convolutional neural network framework, and our image processing module.
After several experiments, we reached the optimal classification accuracy of 93.59% for 57 tree species. Experiments involve 16096 training and 3020 testing leaf images. According to our findings, we come to the following conclusion. Certain image processing procedures for extracting features such as shape and texture descriptors, which we have used in our project, does not produce features as feasible as convolutional neural networks.
"Created by group paY inekereG" => METU Students
Oxirgi yangilanish
Xavfsizlik — dastur ishlab chiquvchilar maʼlumotlaringizni qanday jamlashi va ulashishini tushunishdan boshlanadi. Maʼlumotlar maxfiyligi va xavfsizlik amaliyotlari ilovadan foydalanish, hudud va yoshga qarab farq qilishi mumkin. Bu axborot dastur ishlab chiquvchi tomonidan taqdim etilgan va keyinchalik yangilanishi mumkin.
Reytinglar va sharhlar
2,0
121 ta sharh
Nima yangiliklar
Sunucu güncellemesi, hata düzeltmeleri
