embedded Computer Vision sample video gallery
Aquarium Sobel Filter - eCV on i.MX6
Demonstration of a 2 dimensional Sobel operator used for edge detection. In this demo, the untouched 720p source video is shown on the left with the filtered output on the right.
Multiple Filters - eCV on i.MX6
Demonstration of Multiple Filters operating on single input source
Top Left: Source 720p26
Top Right: SobelXY (2 dimensional Sobel)
Bottom Left: Approximate Median Filter (AMF)
Bottom Right: Morphological Close (erode(dilate(input source))
Background subtraction - eCV on i.MX6
Background Subtraction using Approximate Median Filter (AMF) to segment motion followed mask operation with source image to display only moving objects.
Canny Edge Detection - eCV on i.MX6
Demonstration of real time edge detection using Gaussian Blur for noise reduction followed by Canny Edge Detection. Filtered video in middle segment, source on top and bottom.
Object Tracking - eCV on i.MX6
Object tracking using Approximate Mean Filtering (AMF) for background subtraction followed by Morphology (5x5 close) for blob detection and tracking. Pipeline stages are shown as sequential tiles.
Object Tracking - OpenCV on i.MX6
Object tracking using standard Open CV functions for background subtraction. eCV SDK used to create onscreen graphical overlays.
Sensor Fusion - eCV on i.MX6
Data fusion demo blending real time HD video (720p30) from a standard CMOS sensor with Low Res (32x31) thermal data from infrared sensor and streaming H.264 compressed video over RTSP.