Computer Vision avec Simulink

Track a marker on a whiteboard by using images of the Simulink.ImageType data type.

Run a video processing system on multiple cores using dataflow execution domain in Simulink®.

Use color information to detect and track road edges set in primarily residential settings where lane markings may not be present. The Color-based Tracking example illustrates how to use the Color Space Conversion block, the Hough Transform block, and the Kalman Filter block to detect and track information using hue and saturation.

Detect and track road lane markers in a video sequence and notifies the driver if they are moving across a lane. The example illustrates how to use the Hough Transform, Hough Lines and Kalman Filter blocks to create a line detection and tracking algorithm. The example implements this algorithm using the following steps: 1) Detect lane markers in the current video frame. 2) Match the current lane markers with those detected in the previous video frame. 3) Find the left and right lane markers. 4) Issue a warning message if the vehicle moves across either of the lane markers.

Detect and count cars in a video sequence using Gaussian mixture models (GMMs).

Track a person's face and hand using a color-based segmentation method.

Track objects at a train station and to determine which ones remain stationary. Abandoned objects in public areas concern authorities since they might pose a security risk. Algorithms, such as the one used in this example, can be used to assist security officers monitoring live surveillance video by directing their attention to a potential area of interest.

Detect and track cars in a video sequence using optical flow estimation.

Inspect the concentricity of both the core and the cladding in a cross-section of optical fiber. Concentricity is a measure of how centered the core is within the cladding.

Use sum of absolute differences (SAD) method for detecting motion in a video sequence. This example applies SAD independently to four quadrants of a video sequence. If motion is detected in a quadrant, the example highlights the quadrant in red.

Use the 2-D normalized cross-correlation for pattern matching and target tracking.

Use basic morphological operators to extract information from a video stream. In this case, the model counts the number of staples in each video frame. Note that the focus and lighting change in each video frame.

Segment video in time. The algorithm in this example can be used to detect major changes in video streams, such as when a commercial begins and ends. It can be useful when editing video or when you want to skip ahead through certain content.

Process surveillance video to select frames that contain motion. Security concerns mandate continuous monitoring of important locations using video cameras. To efficiently record, review, and archive this massive amount of data, you can either reduce the video frame size or reduce the total number of video frames you record. This example illustrates the latter approach. In it, motion in the camera's field of view triggers the capture of "interesting" video frames.

Recognize traffic warning signs, such as Stop, Do Not Enter, and Yield, in a color video sequence.

Remove the effect of camera motion from a video stream.

Use the From Video Device block provided by Image Acquisition Toolbox™ to acquire live image data from a Point Grey Flea® 2 camera into Simulink®. The example uses the Computer Vision Toolbox™ to create an image processing system which can recognize and interpret a GTIN-13 barcode. The GTIN-13 barcode, formally known as EAN-13, is an international barcode standard. It is a superset of the widely used UPC standard.

Create a mosaic from a video sequence. Video mosaicking is the process of stitching video frames together to form a comprehensive view of the scene. The resulting mosaic image is a compact representation of the video data. The Video Mosaicking block is often used in video compression and surveillance applications.

Use the From Video Device block provided by Image Acquisition Toolbox™ to acquire live image data from a Hamamatsu C8484 camera into Simulink®. The Prewitt method is applied to find the edges of objects in the input video stream.