At present, the classical semantic segmentation methods can not accurately realize the obstacle instance detection of traffic scene, which makes this method can not be used in driverless system alone. In order to meet the requirements of obstacle detection accuracy in unmanned system, pixel level obstacle detection in complex driving scene is studied in this paper. A pixel level obstacle anomaly detection framework is formed by combining the maximum entropy, maximum distance and perceptual difference of uncertainty mapping with the output of dissimilar model. The framework uses uncertainty map and designs a hollow spatial pyramid pool structure of different receptive field stitching to enhance the correlation between various levels of information to improve the existing re synthesis methods to find the differences between the input image and the generated image. As a general framework, this method focuses on the trained segmentation network to ensure anomaly detection without affecting the accuracy of segmentation. The network is implemented based on pytorch framework. The experimental results on the road scene dataset cityscapes dataset show that the mlou reaches 85.7%.
In the process of actual measurement and analysis of micro near infrared spectrometer, genetic algorithm is used to select the wavelengths and then partial least square method is used for modeling and analyzing. Because genetic algorithm has the disadvantages of slow convergence and difficult parameter setting, and partial least square method in dealing with nonlinear data is far from being satisfactory, the practical application effect of partial least square method based on genetic algorithm is severely affected negatively. The paper introduces the fundamental principles of particle swarm optimization and support vector machine, and proposes a support vector machine method based on particle swarm optimization. The method can overcome the disadvantage of partial least squares method based on genetic algorithm to a certain extent. Finally, the method is tested by an example, and the results show that the method is effective.
KEYWORDS: Video, Motion models, Video surveillance, Statistical modeling, Optical flow, Data modeling, Video processing, Data processing, Detection and tracking algorithms, Feature extraction
Abnormal event detection plays a critical role for intelligent video surveillance, and detection in crowded scenes is a challenging but more practical task. We present an abnormal event detection method for crowded video. Region-wise modeling is proposed to address the inconsistent detected motion of the same object due to different depths of field. Comparing to traditional block-wise modeling, the region-wise method not only can reduce heavily the number of models to be built but also can enrich the samples for training the normal events model. In order to reduce the computational burden and make the region-based anomaly detection feasible, a saliency detection technique is adopted in this paper. By identifying the salient parts of the image sequences, the irrelevant blocks are ignored, which removes the disturbance and improves the detection performance further. Experiments on the benchmark dataset and comparisons with the state-of-the-art algorithms validate the advantages of the proposed method.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.