Real‐time vehicle shadow detection

Abstract

In many traffic-related applications, such as traffic management and structural health monitoring for roads, an accurate estimation of a moving vehicle's size and shape is needed before proceeding further. However, due to the presence of cast shadows, these properties cannot be obtained accurately using common object detection systems. To deal with the problem of misclassifying shadows as foreground, various methods have been introduced. Most of these methods often fail to distinguish shadow points from the foreground object when the boundary between the umbra and the object is unclear due to camouflage. A novel method for detecting moving shadows of vehicles in real-time applications is presented. The method is based on two measurements, namely, the illumination direction and the intensity measurements in the neighbouring pixels in a scanned line. A major advantage of using image lines for classification is the ability to solve the problem associated with camouflages. Experimental results show that the proposed method is efficient in real-time performances and has achieved higher detection rate and discrimination rate when compared with two well-known methods.

Real-time search-free multiple license plate recognition via likelihood estimation of saliency

“…Figure 14: (a) sample image, (b) result of our proposed object detection algortim, (c) result of [17] As shown in Figure 14, the algorithms developed by Wang et al [17] and Liu et al [19] have a high false alarm rate and low accuracy in object detection. Figure 15 shows the outcome of our proposed method comparing to those developed by Russel et al [20] and Zhu et al [33]. As shown in this figure, in [20] some parts of objects are eliminated when the shadow detection and removal algorithm is applied.…”

“…Figure 15 shows the outcome of our proposed method comparing to those developed by Russel et al [20] and Zhu et al [33]. As shown in this figure, in [20] some parts of objects are eliminated when the shadow detection and removal algorithm is applied.…”

“…On the other hand, Wazalwar et al [31] make use of a smoothing operation, detection of edges and dilation operation in their localization algorithm. This paper is inspired by [2,11,14,20,31] and we extend our previous work [9,10] combined with Bayesian approach used in [27,28] to estimate the 3-D salient regions by exploiting the motion information in the videos. The proposed algorithm is fast and robust when used in the changing environment and is suitable for localizing multiple plates.…”

“…Russell et al [20] also proposed a simple method for detecting and removing shadows in traffic videos by using spatial intensity relationship between pixels associated to a scanned image line. This method performs the classification on image lines to differentiate between the background and the shadow and has a high accuracy rate.…”

Real-time detection of moving cast shadows using foreground luminance statistics

“…However, some methods are suited only for specific environments (indoor/outdoor) and specific targets (human/vehicle/others). [15][16][17] For instance, Ref. [17] only deals with detection of vehicle cast shadows.…”