Blob recognition from camera images

I have a C# program that uses DIRECTX to retrieve images from a webcam, converts them to monochrome,? and identifies the biggest blob in the images in real-time.

You can download the attached files and try it with any webcam. Upon launching the program, you have to first select a video device as an input source. Then you can click on the "Toggle Display" button to see? processed images. You can adjust the parameters "hover position" and "hover height" to change the size of the processing window. The intersection of the 2 red lines identifies the center of the biggest blob the program can find in the images. The FPS number is the frame rate.

Problem: With a regular camera, the frame rate can go up to more than 20 FPS. However, if I put on the camera a filter for infrared light only, the frame rate reduces significantly to under 8 FPS, while the CPU utilization is still very low (around 10%). I need help to? identify the root cause of the problem and to modify the? code to increase the frame rate to near 30 FPS.

Upon starting the project, I will lend you a webcam with a built-in infrared filter. I will also send you the source code (excluding the part which identifies? image blobs).

I'm looking for local coders only, within 40 miles of zip code 20109, VA, USA.

Habilidades: Programação C, Programação C#, Engenharia, Microsoft, MySQL, PHP, Gestão de projetos, Arquitetura de software, Teste de Software, Área de trabalho do Windows

Ver mais: usa will need coders, usa programming, programming center, php programming in video download, on the intersection, near c programming, lines intersection, intersection of lines, intersection of 3 lines, intersection of 2 lines, intersection and, intersection 2 lines, find the intersection, find intersection of 2 lines, find intersection, find coders near you, find c+ coders, find an intersection, directx programming, webcam camera, red camera, number recognition, infrared, fps, find root

Acerca do Empregador:
( 2 comentários ) United States

ID do Projeto: #3020388