Facial detection
AssigneeAnalog Devices Global
InventorsAnil M. Sripadarao, Bijesh Poyil
Life of this patent
Estimating the term…
Abstract
In one aspect, there is disclosed a digital signal processor and method performed by the same for performing object detection, including facial detection, in a reduced number of clock cycles. The method comprises using Sobel edge detection to identify regions with many edges, and classifying those regions as foreground candidates. Foreground candidates are further checked for vertical or horizontal symmetry, and symmetrical windows are classified as face candidates. Viola-Jones type facial detection is then performed only on those windows identified as face candidates.
Claims21
An object detection method, performed on a digital signal processor, for identifying objects of interest in a foreground region of a digital image, the method comprising:
- dividing the digital image into a plurality of windows;
- performing edge detection on each window;
- classifying each window as a foreground candidate or background candidate, wherein classifying comprises computing a number of edges for the window and classifying the window as a foreground candidate only if the number of edges for the window is greater than a threshold;
- and
- performing additional object detection only on those windows classified as foreground candidates.
A tangible non-transitory computer-readable storage medium having stored thereon software instructions for a digital signal processor, the instructions adapted for identifying objects of interest in a foreground region of a digital image, the instructions configured to instruct a processor to:
- divide the digital image into a plurality of windows;
- perform edge detection on each window;
- classify each window as a foreground candidate or background candidate, wherein classifying comprises computing a number of edges for the window and classifying the window as a foreground candidate only if the number of edges for the window is greater than a threshold;
- and
- perform additional object detection only on those windows classified as foreground candidates.
A digital signal processing system configured to identify objects of interest in a foreground region of a digital image with reduced consumption of clock cycles, the system comprising:
- a processor;
- and
- a memory communicatively coupled to the processor and having stored thereon instructions that, when executed, instruct the processor to:
- divide the digital image into a plurality of windows;
- perform edge detection on each window;
- classify each window as a foreground candidate or background candidate, wherein classifying comprises computing a number of edges for the window and classifying the window as a foreground candidate only if the number of edges for the window is greater than a threshold;
- and
- perform additional object detection only on those windows classified as foreground candidates.
Dependencies are inferred from the claim text.