A method for zip code recognition is presented. 2D binary images are input to HAVNET, a neural network which employs the Hausdorff distance as a similarity metric to train the weights which are required to represent t...
详细信息
ISBN:
(纸本)0769503187
A method for zip code recognition is presented. 2D binary images are input to HAVNET, a neural network which employs the Hausdorff distance as a similarity metric to train the weights which are required to represent the patterns learned by the network. A new learning rule for HAVNET is also introduced. In this approach, HAVNET is combined with a multilayer neural network. The Hausdorff distance acts as a zip code filter. Only the confusing digits are input into the multilayer network for training or recognition. Experimental results show that the correct-recognition rate with zero rejection is more than 97% for a database used by the Chinese mail sorting system.
The activation of ancient vessel-fragments is a time consuming but important task for archaeologists. The basis for classification and reconstruction is the profile which is the cross-section of the fragment in the di...
详细信息
The activation of ancient vessel-fragments is a time consuming but important task for archaeologists. The basis for classification and reconstruction is the profile which is the cross-section of the fragment in the direction of the rotational axis of symmetry. Hence the position of a fragment (orientation) on a vessel is important. In this work the estimation of the axis of rotation out of range data by using a Hough inspired method is proposed. In order to avoid outliers a robust method for estimation of the axis is used. Classification and reconstruction are performed in a bottom-up manner using a description language, which holds all features of the fragment as primitives and all properties among features as relations. Classification of newly found fragments of unknown type is performed by comparing the description of the new fragment with the description of already classified fragments by completing graph similarity. The sub-graph with the highest similarity is then used to reconstruct the complete vessel out of the fragment.
The eigendecomposition of banded symmetric matrices Is important in regularized signal restoration. In this paper a new fast decomposition algorithm is developed by using FFT and rank-one update. In this way the split...
详细信息
The eigendecomposition of banded symmetric matrices Is important in regularized signal restoration. In this paper a new fast decomposition algorithm is developed by using FFT and rank-one update. In this way the split of the matrix becomes more direct and the cost of the initial decomposition decreases. The algorithm is especially useful for parallel processing. Two numerical examples are given, which show that the new method can achieve comparable results to the classical methods.
A practical postal numeral segmentation and recognition system for Chinese business letters is presented. Line information for the address blocks is gained from the envelope image by projection, then the sub-image of ...
详细信息
ISBN:
(纸本)0769503187
A practical postal numeral segmentation and recognition system for Chinese business letters is presented. Line information for the address blocks is gained from the envelope image by projection, then the sub-image of the postal numerals is segmented. The numerals are recognized by three independent classifiers, and their results are combined to form final results. An improved voting principle based on a priori knowledge of each classifier is utilized to combine multiple classifiers in order to achieve perfect performance for high recognition rate and low error rate. The recognition ability of each classifier, acting with the voting principle, is obtained by statistics for a large set of samples. Different thresholds gained by training are employed for different classes, which can yield better results. A parallel structure hardware is designed to implement a real-time imageprocessing system. The performance of the system is shown to be satisfactory, sufficient for the requirements of practical use.
To improve the reliability of a template-matching classifier for recognizing connected handwritten characters, we present an evolutionary algorithm to optimize handwritten numeral templates represented by rational B-s...
详细信息
Noise removal is an important problem in many applications. In this paper a new two-step scheme of the decision-based impulse noise removal method by means of contaminated pixel detection is proposed and comparison wi...
详细信息
Noise removal is an important problem in many applications. In this paper a new two-step scheme of the decision-based impulse noise removal method by means of contaminated pixel detection is proposed and comparison wi...
详细信息
Noise removal is an important problem in many applications. In this paper a new two-step scheme of the decision-based impulse noise removal method by means of contaminated pixel detection is proposed and comparison with direct order statistic filtering is given. The proposed methods satisfy both objective and subjective image quality.
A new edge detection operator based on image feature is proposed,which analyze edges in image for edge feature in two *** local extreme of the operator is created at the edge location and low value is created at the s...
详细信息
A new edge detection operator based on image feature is proposed,which analyze edges in image for edge feature in two *** local extreme of the operator is created at the edge location and low value is created at the smooth *** can be located by obtaining the local extreme and a threshold of the operator response. The detection operator is shown to be better than Canny operator in terms of signal-to-noise ratio and edge location accuracy.
This paper presents an application of 3D-reconstruction and graph theory in the field of archaeology. The classification and reconstruction of ancient pots and vessels out of fragments (so-called sherds) is an importa...
详细信息
This paper presents an application of 3D-reconstruction and graph theory in the field of archaeology. The classification and reconstruction of ancient pots and vessels out of fragments (so-called sherds) is an important aspect of archaeological research work. Up to now this is a time consuming, inaccurate, and subjective task which leads to tons of unclassified fragments in archives. Computer aided classification could help to get a better understanding of ancient cultures, since all data of an excavation would be accessible to the public, not only- selected parts as it is now. We propose a bottom-up strategy to classify fragments. The profile section (which is a section a the fragment in the direction of the rotational axis) is segmented into its primitives (with certain properties like length) and relations among this primitives (like position and curvature of connecting points). These primitives and the relations form a description language, different profiles have different descriptions.
A new edge detection operator based on image features is proposed, which analyzes edges in images for edge features in two dimensions. The local extreme of the operator is created at the edge location and a low value ...
详细信息
A new edge detection operator based on image features is proposed, which analyzes edges in images for edge features in two dimensions. The local extreme of the operator is created at the edge location and a low value is created at the smooth region. Edges can be located by obtaining the local extreme and a threshold of the operator response. The detection operator is shown to be better than the Canny operator in terms of signal-to-noise ratio and edge location accuracy.
暂无评论