Shadow detection using colour and edge information

Shadows appear in many scenes. Human can easily distinguish shadows from objects, but it is one of the challenges for Shadow Detection Intelligent Automated Systems. Accurate shadow detection can be difficult due to the illumination variations of the background and similarity between appearance of the objects and the background. Colour and edge information are two popular features that have been used to distinguish cast shadows from objects. Colour information is useful because information such as hue in HSI colour model, Y in YCbCr colour model, the gradient of red, green and blue channels in RGB colour model are invariant in both shadow area and background, but information like intensity is different. Besides, the useful information for shadow detection is the cast shadow that does not have exterior edges. However, this become a problem when the difference of colour information between object, shadow and background is poor, the edge of the shadow area is not clear and the shadow detection method is supposed to use only for colour or edge information method. In this research,a shadow detection method using both colour and edge information is presented. As a result, in the absence of colour information, the edge information is used and in the absence of edge information, the colour information is used. Shadow pixels are detected based on the colour information (using YCbCr, HSI, extended c1c2c3 and hue difference of foreground and background). In order to improve the accuracy of shadow detection using colour information, a new formula is used in the denominator of original c1c2c3. In addition using the hue difference of foreground and background is proposed. Furthermore, edge information is applied separately and the results are combined using a Boolean operator (logical AND). In order to evaluate the performance of the proposed method, Shadow Detection Rate, Shadow Discrimination Rate, and Fscore from the extracted shadow image are computed. The above-mentioned factors are calculated and compared with each other in the following conditions namely detection using colour information method with different colour features, edge information method, and combination of these two methods. The experiments were done using VC++ 2008 with different standard indoor and outdoor data sets. These experiments investigate the performance of the proposed method in comparison with the Bangyu’s method and Panicker’s method which are based on colour and edge information. The results show the accuracy of detected shadow pixels is improved to 10%.

Preview PDF FSKTM 2013 4R.pdf Download (920kB) | Preview

Actions (login required)

View Item