Traffic light is one of the important signs for drivers that help managing the car flow and reducing accident on the road. As of today technology, there exists a traffic light detection system that warns the driver to reduce the accident significantly. In this paper, we are concerned with only the red and yellow traffic light to reduce false positive and time consumption. The fast radial symmetry transform is a fast variation of the circular Hough transform. The CIELab color model is used because it is less dependent on lighting conditions. The precision and the recall of the proposed method are increased from previous work.

This paper presents a drowsiness detection for drivers. The drowsiness is detected by monitoring the eye state (open or close). Firstly, human face detection is performed using the Haar cascade method. Within the facial region, we can approximately determine the eye regions, further restricting the region of interest (ROI). We then locate a dark circular object using two vectors: one is distance vectors and the other gradient vectors. The cross-correlation between these two vectors should be maximized at the center of a dark circle. Experimental results show that the proposed method works well in both bright and dark conditions. The computation speed of the proposed method is fast enough to perform at a video rate.

Crowd counting is a challenging task dealing with the variation of an object scale and a crowd density. Existing works have emphasized on skip connections by integrating shallower layers with deeper layers, where each layer extracts features in a different object scale and crowd density. However, only high-level features are emphasized while ignoring low-level features. This paper proposes an estimation network by passing high-level features to shallow layers and emphasizing its low-level feature. Since an estimation network is a hierarchical network, a high-level feature is also emphasized by an improved low-level feature. Our estimation network consists of two identical networks for extracting a high-level feature and estimating the final result. To preserve semantic information, dilated convolution is employed without resizing the feature map. Our method was tested in three datasets for counting humans and vehicles in a crowd image. The counting performance is evaluated by mean absolute error and root mean squared error indicating the accuracy and robustness of an estimation network, respectively. The experimental result shows that our network outperforms other related works in a high crowd density and is effective for reducing over-counting error in the overall case.

No abstract