Object detection in images
AssigneeBritish Telecommunications plc
InventorsLi-Qun Xu, Jose-Luis Landabaso
Life of this patent
Estimating the term…
Abstract
A segmentation operation is applied to an input image to identify foreground objects of interest, and then a shadow removal operation is applied to remove any detected shadows from the foreground segmentation. The shadow removal algorithms can leave holes and bisections in the segmentation map, however, which will then subsequently impact on an object detection step performed using connected component analysis. To get around this problem, a conditional morphological dilation operation is applied to the segmentation map to ‘grow’ the segmented blobs to fill in any holes and bisections, without re-growing shadow pixels in the segmentation. The result is an object detection method and system which is robust to illumination changes causing shadows and/or highlights.
Claims5
An image processing method for detecting objects within an input image, the image being composed of picture elements, the method comprising:
- a) segmenting picture elements representing a foreground object within the input image from those picture elements forming the image background using a first segmentation technique, wherein the picture elements segmented as foreground include elements representing a shadow or highlight of the object;
- b) storing a segmentation mask, containing the location of the segmented picture elements representing a foreground object, in a data store;
- c) segmenting picture elements which have the characteristics of a shadow or highlight of an object from those picture elements representing the foreground object using at least one other segmentation technique adapted to detect shadows or highlights;
- d) segmenting as foreground surrounding picture elements to those picture elements which are already segmented as foreground by performing a morphological dilation operation;
- e) comparing the surrounding picture element segmented as foreground in step d) against the stored segmentation mask;
- f) repeating steps d) and e) if comparison determines that the segmented surrounding picture elements do not touch the boundary of the foreground object in the stored segmentation mask;
- and
- g) if the comparison determines that the segmented surrounding picture elements touch the boundary of the foreground object in the stored segmentation mask, then detecting as objects, groups of adjacent picture elements which have been segmented as foreground.
An image processing system for detecting objects within an input image, the image being composed of picture elements, the system comprising:
- a data store;
- an image processor arranged to receive an input image to be processed, and to apply the following image processing operations thereto:
- a) to segment picture elements representing a foreground or moving object within the input image from those picture elements forming the image background using a first segmentation technique adapted to detect differences in the input image from a general background image, wherein the picture elements segmented as foreground include elements representing a shadow or highlight of the object;
- b) to store a segmentation mask, containing the location of the segmented picture elements representing a foreground object in the data store;
- c) to segment picture elements which have the characteristics of a shadow or highlight of an object from those picture elements representing the foreground object using at least one other segmentation technique adapted to detect shadows or highlights;
- d) to repeatedly segment as foreground surrounding picture elements to those picture elements already segmented as foreground using a morphological dilation operation
- e) to compare the surrounding picture elements segmented as foreground in step d) against the stored segmentation mask;
- f) to stop the morphological dilation operation if the comparison determines that segmented surrounding picture elements touch the boundary of the foreground object in the stored segmentation mask;
- and then
- g) to detect as objects, groups of adjacent picture elements which have been segmented as foreground.
Dependencies are inferred from the claim text.