Key Features
- Advanced object classification core for support of Vehicle Detection in camera-based video systems
- Supports AMD Zynq™ 7000 SoC and Zynq™ UltraScale+™ MPSoC
- ARM® AMBA® AXI4-Stream Slave compliant video input
- AXI4-Lite compliant registers interface
- Supports resolutions up to 4096x4096
- Support for multiple scale detection
- Template size 24x24 pixels
- Run-time variable window shift
- High Input Data Rate; > 150 Mpixel/sec
- High Throughput; > 64 GOP/sec
- Frame rate of 25 fps with multiple scales
- Detection range from 5 to 100m
- C code API and post-processing library available
- Prepackaged for the AMD Vivado™ Design Suite
Description
The logiVDET is a learning-based vehicle detection IP core, developed for vision-based embedded applications. The algorithm follows a discriminative approach based on a cascaded classifier using Local Binary Pattern features. This architecture makes the detection process faster by rejecting the negative examples in the initial stages of the cascade, while the computation effort is mainly spent on the templates hard to classify. The IP core works at a single scale, i.e. it recognizes vehicles at a fixed size. Extension to multiple scales is given by inserting the core in a framework that provides it with a sequence of re-scaled versions of the same input frame. This way, it is possible to detect vehicles in an arbitrary range of distances. For example, with 20 levels of scale (1 MPixel camera with 50 degrees of lens FOV) it is possible to detect vehicles in a range from 5 to 100 m running at 30 fps.
| The logiVDET core is sourced from Technology Partner
EVS Embedded Vision Systems Srl. |