This paper focuses on route planning, especially for unmanned aircrafts in marine environment. Firstly, new heuristic information is adopted such as threat-zone, turn maneuver and forbid-zone based on voyage heuristic...
详细信息
This paper focuses on route planning, especially for unmanned aircrafts in marine environment. Firstly, new heuristic information is adopted such as threat-zone, turn maneuver and forbid-zone based on voyage heuristic information. Then, the cost function is normalized to obtain more flexible and reasonable routes. Finally, an improved sparse A* search algorithm is employed to enhance the planning efficiency and reduce the planning time. Experiment results showed that the improved algorithm for aircraft in maritime environment could find a combinational optimum route quickly, which detoured threat-zones, with fewer turn maneuver, totally avoiding forbid-zones, and shorter voyage.
This paper presents a novel adaptive tracking algorithm that fuses multiple cues based on feature uncertainty measurement in the particle filter framework. We first introduce a self-adaptive multi-cue fusion strategy,...
详细信息
ISBN:
(纸本)9781424496297
This paper presents a novel adaptive tracking algorithm that fuses multiple cues based on feature uncertainty measurement in the particle filter framework. We first introduce a self-adaptive multi-cue fusion strategy, which overcomes the drawbacks of the traditional product fusion and sum fusion strategies. Furthermore, the proposed strategy effectively sharpens the distribution of the fused posterior as well as makes the tracking results less sensitive to the noise. Then, based on the fact that tracking failure often happens in the cases of low discriminative abilities of the observed features, we define a new feature uncertainty measurement. The proposed uncertainty measurement is thereafter used to adaptively adjust the relative contributions of different cues to tracking. An extensive number of comparative experiments show that the proposed tracking algorithm is more stable and robust than the single feature, product fusion, and sum fusion tracking algorithms.
In the patternrecognition subspace method, the researcher has paid more attention to extract feature subspace, then expressed individual prototype with the training sample mean. Because the number of training sample ...
详细信息
In the patternrecognition subspace method, the researcher has paid more attention to extract feature subspace, then expressed individual prototype with the training sample mean. Because the number of training sample is limited, there is certain difference between the sample mean and the individual prototype. In order to reduce this difference, a sample restraint clustering algorithm was proposed, which make up of it's clustering objective function with the Fisher criterion, and it's goal lies in minimizing within classes and maximizing between classes. The recurrence formula computing each cluster centroid is derived direct from the objective function. In the random produced sample space, the clustering experiment indicated the proposed method is able to reduce disparity between cluster centroid and the individual prototype. In the face recognition experiment, the positive recognition ratio of some algorithms may be improved when it's prototype is replaced with cluster centroid calculated by the proposed algorithm rather than the mean of training samples.
This paper presents a new algorithm for Braille cells recognition using image processing technique. Scanned Braille document is composed from three classes of gray-level: (i) background, (ii) recto dots, and (iii) ver...
详细信息
This paper presents a new algorithm for Braille cells recognition using image processing technique. Scanned Braille document is composed from three classes of gray-level: (i) background, (ii) recto dots, and (iii) verso dots. We segment the Braille image using a stability thresholding method with a mixture of Beta distributions. To ensure correct detection and extraction of dots composing Braille cells, a grid is formed to contain the Braille dots. We identified a recto dot by a light region that exists above a dark region using the segmented image. In the same way, we identify a verso dot in double sided document by a light region that exists below a dark region. After having recto and verso dots, Braille cells are then recognized based on the standard regrouping of dots. Experimentation showed that Braille cells composing are automatically identified from those grids with excellent accuracy.
Basic: With the proposal of the smart grid strategy, there are higher requirement for the real time dispatching and unit commitment, objectively, the new generation of unit commitment should be dynamic, smart, refinin...
详细信息
Basic: With the proposal of the smart grid strategy, there are higher requirement for the real time dispatching and unit commitment, objectively, the new generation of unit commitment should be dynamic, smart, refining efficiency and operable. The multi-objective unit commitment intelligent optimization system for unit commitment based on the power grid state recognition which this paper established is an improvement of the existing unit commitment optimization pattern. It make use of knowledge base system to it identify and classify the real time operation state of power grid, base on which it determines “the primary optimization objectives” and “the primary constraints”, meanwhile it achieves the simplification and dynamic modeling of the multi-objective problem. It does not only improve the accuracy of the multi-target model and make the model targeted, but also simplify the model and improve the computing efficiency.
Data-centric affect modeling may render itself restrictive in practical applications for three reasons, namely, it falls short of feature optimization, infers discrete affect classes, and deals with relatively small t...
详细信息
ISBN:
(纸本)9781605585154
Data-centric affect modeling may render itself restrictive in practical applications for three reasons, namely, it falls short of feature optimization, infers discrete affect classes, and deals with relatively small to average sized datasets. Though it seems practical to use the feature combinations already associated to commonly investigated sensors, there may be other potentially optimal features that can lead to new relations. Secondly, although it seems more realistic to view affect as continuous, it requires using continuous labels that will increase the difficulty of modeling. Lastly, although a large scale dataset reflects a more precise range of values for any given feature, it severely hinders computational efficiency. We address these problems when inferring physiology-affect relations from datasets that contain 2-3 million feature vectors, each with 49 features and labelled with continuous affect values. We employ automatic feature selection to acquire near optimal feature subsets and a fast approximate kNN algorithm to solve the regression problem and cope with the challenge of a large scale dataset. Our results show that high estimation accuracy may be achieved even when the selected feature subset is only about 7% of the original features. May the results here motivate the HCI community to pursue affect modeling without being deterred by large datasets and further the discussions on acquiring optimal features for accurate continuous affect approximation.
In this research a hybrid feature selection technique based on genetic and simulated annealing algorithms is proposed. this approach is evaluated by using Bayesian classifier on a dataset of handprinted Farsi characte...
详细信息
In this research a hybrid feature selection technique based on genetic and simulated annealing algorithms is proposed. this approach is evaluated by using Bayesian classifier on a dataset of handprinted Farsi characters which includes 100 samples for each 33 hand-printed characters. The acquired results have been improved by correction of Simulated Annealing through considering two minimum and maximum thresholds.
A novel algorithm is presented for detecting a target placed in background and degraded by non-uniform illumination and additive noise. The algorithm includes two steps. First, the input scene is preprocessed with est...
详细信息
ISBN:
(纸本)9781424465828
A novel algorithm is presented for detecting a target placed in background and degraded by non-uniform illumination and additive noise. The algorithm includes two steps. First, the input scene is preprocessed with estimated illumination function. Then the processed scene is correlated with a designed optimum filter based on Wiener filtering. Simulation results show effectiveness of the proposed method in terms of the robustness to non-uniform illumination and noisy conditions.
The QMS (Quantum Morphogenetic System) is a mathematical model devised to give a different more elegant and intuitive perspective of the pre-processing by assuming a non-Euclidean geometry. Input image is projected to...
详细信息
The QMS (Quantum Morphogenetic System) is a mathematical model devised to give a different more elegant and intuitive perspective of the pre-processing by assuming a non-Euclidean geometry. Input image is projected to the feature vector in terms of its contra-variant components which are interdependent. Image reconstruction is achieved by parallelogram summation of the vector of those components. View-Invariant recognition is achieved through the manipulation of association matrix which in geometrical perspective is a deformation of space towards a canonical point for all vectors which are being considered the same.
Conventional Gabor filter-based approaches are global considerations for the selections of Gabor filters. Owing to fingerprint patterns are full of ridges and valleys; the whole fingerprint can be viewed as the compos...
详细信息
Conventional Gabor filter-based approaches are global considerations for the selections of Gabor filters. Owing to fingerprint patterns are full of ridges and valleys; the whole fingerprint can be viewed as the composition of many various local ridge-valley structures. Based on the various local ridge orientations and frequencies, the whole fingerprint is represented by a template of local Gabor filters. Therefore, the core point detection and fingerprint matching are directly performed from gray-scale images without preprocessing. As compared with the global Gabor filter-based approach, the proposed method not only saves memory space, but also speeds up the whole procedure. Testing with the same database, higher recognition rates are also achieved for the local approach.
暂无评论