Pedestrian detection is a critical feature of autonomous vehicle or advanced driver assistance system. Thispaper presents a novel instrument for pedestrian detection bycombining stereo vision cameras with a thermal camera. A newdataset for vehicle applications is built from the test vehiclerecorded data when driving on city roads. Data received frommultiple cameras are aligned using trifocal tensor with precalibrated parameters. Candidates are generated from eachimage frame using sliding windows across multiple scales. Areconfigurable detector framework is proposed, in which feature extraction and classification are two separate stages. Theinput to the detector can be the color image, disparity map,thermal data, or any of their combinations. When applying toconvolutional channel features, feature extraction utilizes the firstthree convolutional layers of a pre-trained convolutional neuralnetwork cascaded with an AdaBoost classifier. The evaluationresults show that it significantly outperforms the traditionalhistogram of oriented gradients features. The proposed pedestrian detector with multi-spectral cameras can achieve 9% logaverage miss rate. The experimental dataset is made available athttp://computing.wpi.edu/dataset.html.
To View the Abstract Contents
Now it is Your Time to Shine.
Great careers Start Here.
We Guide you to Every Step
Success! You're Awesome
Thank you for filling out your information!
We’ve sent you an email with your Final Year Project PPT file download link at the email address you provided. Please enjoy, and let us know if there’s anything else we can help you with.
To know more details Call 900 31 31 555
The WISEN Team