پروژه پردازش تصویر sequential ransac" based on SIFT to detect multiple templates in a scene.
شش سال پیش منتشر شده
تعداد بازدید: 430
کد پروژه: 164806
شرح پروژه
Can you help me with the following task?
You should implement "sequential ransac" based on SIFT to detect multiple templates in a scene. The basic steps are feature matching -> ratio test -> clustering of matched features through RANSAC and homographies -> every time you match an object, you should remove the corresponding matches and proceed on the remaining.
Codes should be properly engineered and developed in Matlab. You can use any alternative feature provided in Matlab (e.g. SURF, FREAK).
You can then choose a different feature (e.g. SURF, FREAK...) and compare the detection performance on a couple of scenes. Remember that it has to be a multi-instance detection problem. This means that the scene should contain multiple instances of the template object, the template should be planar and you should be given a single image of that.
Maybe you can anticipate images you plan using and I'll let you know if they are ok
مثلا میتونید در مورد اجرا و دو صفحه گزارش از مقاله زیر براورد کنید یا هر مقاله ای که خودتون پیشنهاد می کنید:
https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=8346440
مهارت ها و تخصص های مورد نیاز
مهلت برای انجام
3روز
وضعیت مناقصه
بسته
درباره کارفرما
عضویت هفت سال پیش
