FAST Corner Detection

Features from accelerated segment test (FAST) is a corner detection algorithm, that is faster than most of the other feature detectors.

ThexFFASTfunction picks up a pixel in the image and compares the intensity of 16 pixels in its neighborhood on a circle, called the Bresenham's circle. If the intensity of 9 contiguous pixels is found to be either more than or less than that of the candidate pixel by a given threshold, then the pixel is declared as a corner. Once the corners are detected, the non-maximal suppression is applied to remove the weaker corners.

This function can be used for both still images and videos. Initially, you have to specify the maximum number of corners. The total number of corners returned will never exceed that maximum value. If the actual number of corners in the image is more than that specified , the threshold can be increased to reduce the number of corners.

API Syntax

template void xFFAST(xF::Mat & _src_mat,ap_uint<32> list[MAXPNTS],unsigned char _threshold,uint32_t *nCorners)

Parameter Descriptions

The following table describes the template and the function parameters.

Table 1.xFFAST Function Parameter Descriptions
Parameter Description
NMS If NMS == 1, non-maximum suppression is applied to detected corners (keypoints). The value should be 0 or 1.
MAXPNTS Maximum number of corners that can be detected by the kernel.
SRC_T Input pixel type. Only 8-bit, unsigned, 1-channel is supported (XF_8UC1)
ROWS Maximum height of input image (must be a multiple of 8)
COLS Maximum width of input image (must be a multiple of 8)
NPC Number of pixels to be processed per cycle; possible options are XF_NPPC1 and XF_NPPC8 for 1 pixel and 8 pixel operations respectively.
_src_mat Input image
list List of corners. The corners are packed in 32-bit format. The lower 16-bits provides the column index and the upper 16-bits indicates the row index.
_threshold Threshold on the intensity difference between the center pixel and its neighbors. Usually it is taken around 20.
nCorners The number of corners detected in the input image, which is the output of kernel.

Resource Utilization

The following table summarizes the resource utilization of the kernel for different configurations, generated usingVivado HLS 2017.1for theXilinx Xczu9eg-ffvb1156-1-i-es1FPGA, to process a grayscale HD (1080x1920) image for 1024 corners with NMS.

Table 2.xFFAST Function Resource Utilization Summary

Name

 Resource Utilization
1 pixel 8 pixel
300 MHz 150 MHz
BRAM_18K 10 28
DSP48E 0 0
FF 2695 7310
LUT 3792 20956
CLB 769 3519

Performance Estimate

The following table summarizes the performance of kernel for different configurations, as generated usingVivado HLS 2017.1tool for theXilinx Xczu9eg-ffvb1156-1-i-es1, to process a grayscale HD (1080x1920) image for 1024 corners with non-maximum suppression (NMS).

Table 3.xFFAST Function Performance Estimate Summary
Operating Mode

Operating Frequency

(MHz)

 Filter Size Latency Estimate
Max (ms)
1 pixel 300 3x3 7
8 pixel 150 3x3 1.86