Capra Image Processing Platform

Capra Image Processing Platform is an image processing framework which acquires images from different sources and allows real time image processing through a pipeline which can be configured in accordance with the requirements of specific applications. CAPRA provides users with real time image processing ability without having expertise in this discipline. Our image processing framework Capra also enables users who do not employ software engineers specialized in image processing to provide faster and customized solutions to their customers. CAPRA will provide effective solutions in defense, security and medical imaging sectors besides all industrial sectors which require quality control and surveillance.

As a result of the developments in CPU and GPU technologies, multi-core programming on CPUs and General Purpose GPU Programming (GPGPU) are becoming more widespread today and force software developers to adopt their products to parallel programming which leads to achievement of higher computational potentials. In parallel with these developments, conventional image processing applications, which were created on single CPUs, are being replaced by multi-core programming. This transformation leads to new applications and business opportunities in several sectors including defense, aviation, medical imaging, automotive, textile and civil security.

An image processing framework which is expected to work with high performance in various applications with different requirements should be independent from processor architectures. GPUs and CPUs are the main units used for parallelization. Capra’s design is compatible with all processor architectures and it meets the performance requirements of all sectors it is currently used in since it is built on a pipeline architecture which exploits performance advantages of CPUs, GPUs and FPGAs.

We continue enriching Capra Image Processing Platform with new abilities determined in accordance with our agenda and customer demands. Some filters already included in Capra are:

  • Sensor Fusion
  • Video stabilization
  • Noise elimination
  • Background modeling
  • Fog elimination
  • Snow elimination
  • Target detection
  • Long range motion detection
  • Object tracking
  • Motion detection
  • Fence intrusion detection
  • Suspicious and unattended package detection
  • Suspicious behavior detection
  • Loss of valuable objects
  • Object counting
  • Defect detection
  • People counting
  • Behaviour analysis
  • Eye tracking
  • License plate recognition
  • Face recognition
  • Vehicle counting
  • Traffic density monitoring

Technical features;

  • H.264, MPEG 1, MPEG 2, MPEG 4, JPEG and JPEG 2000 compression
  • Compatible with Windows NT, Windows CE, Linux, MontaVista Linux 6, QNX, Mac OS and Unix/Solaris
  • Fast image processing up to 100 fps, depending on the technical features of cameras
  • Supported inputs;
  • Frame Grabber (PAL-NTSC)
  • Video (MPEG2, MPEG4, H263, H264)
  • Image series (JPG, JPEG2000, PNG etc.)
  • HD
  • Camera Link, LVDS, IEEE1394
  • Analog (PAL, CCIR) and digital (HD, LVDS) input/output.
  • Compatible with thermal and EO cameras