Shadow and variable illumination considerably influence the results of image understanding such as image segmentation, object tracking, and object recognition. The intrinsic image decomposition is to separate the refl...Shadow and variable illumination considerably influence the results of image understanding such as image segmentation, object tracking, and object recognition. The intrinsic image decomposition is to separate the reflectance and the illumination image from an observed image. The intrinsic image decomposition is very useful to remove shadows and then improve the performance of image understanding. In this paper, we present a new shadow removal method based on intrinsic image decomposition on a single color image using the Fisher Linear Discriminant (FLD). Under the assumptions-Lambertian surfaces, approximately Planckian lighting, and narrowband camera sensors, there exist an invariant image, which is 1-dimensional greyscale and independent of illuminant color and intensity. The Fisher Linear Discriminant is applied to create the invariant image. And further the shadows can be removed through the difference between invariant image and original color image. The experimental results on real data show good performance of this algorithm.展开更多
A Prioritized Medium Access Control (P-MAC) protocol is proposed for wireless routers of mesh networks with quality-of-service provisioning. The simple yet effective design of P-MAC offers strict service differentia...A Prioritized Medium Access Control (P-MAC) protocol is proposed for wireless routers of mesh networks with quality-of-service provisioning. The simple yet effective design of P-MAC offers strict service differentiation for prioritized packets. A Markov model is developed to yield important performance matrices including the packet blocking probability due to queue overflow and the packet reneging probability due to delay bound. It is further proved that the service time of P-MAC approximates exponential distribution, and can be effectively estimated. The analytic models with preemptive and non-preemptive schemes, validated via simulations, show that P-MAC can effectively support traffic differentiation and achieve very low packet dropping (both reneging and blocking) probabilities when the traffic load is below the channel capacity. When the network is overloaded, P-MAC can still maintain extremely stable and high channel throughput. Moreover, it is demonstrated that P-MAC performs superior in multihop networks, further proving the advantages of the proposed protocol.展开更多
文摘Shadow and variable illumination considerably influence the results of image understanding such as image segmentation, object tracking, and object recognition. The intrinsic image decomposition is to separate the reflectance and the illumination image from an observed image. The intrinsic image decomposition is very useful to remove shadows and then improve the performance of image understanding. In this paper, we present a new shadow removal method based on intrinsic image decomposition on a single color image using the Fisher Linear Discriminant (FLD). Under the assumptions-Lambertian surfaces, approximately Planckian lighting, and narrowband camera sensors, there exist an invariant image, which is 1-dimensional greyscale and independent of illuminant color and intensity. The Fisher Linear Discriminant is applied to create the invariant image. And further the shadows can be removed through the difference between invariant image and original color image. The experimental results on real data show good performance of this algorithm.
基金Supported in part by the National Science Foundation CAREER Award (No. CNS-0347686)US Department of Energy (DoE) (No. DE-FG02-04ER46136)
文摘A Prioritized Medium Access Control (P-MAC) protocol is proposed for wireless routers of mesh networks with quality-of-service provisioning. The simple yet effective design of P-MAC offers strict service differentiation for prioritized packets. A Markov model is developed to yield important performance matrices including the packet blocking probability due to queue overflow and the packet reneging probability due to delay bound. It is further proved that the service time of P-MAC approximates exponential distribution, and can be effectively estimated. The analytic models with preemptive and non-preemptive schemes, validated via simulations, show that P-MAC can effectively support traffic differentiation and achieve very low packet dropping (both reneging and blocking) probabilities when the traffic load is below the channel capacity. When the network is overloaded, P-MAC can still maintain extremely stable and high channel throughput. Moreover, it is demonstrated that P-MAC performs superior in multihop networks, further proving the advantages of the proposed protocol.