A new method for DNA microarray image segmentation based on patternrecognition techniques is introduced. the method performs an unsupervised classification of pixels using a clustering algorithm, and a subsequent sup...
详细信息
ISBN:
(纸本)9783642040306
A new method for DNA microarray image segmentation based on patternrecognition techniques is introduced. the method performs an unsupervised classification of pixels using a clustering algorithm, and a subsequent supervised classification of the resulting regions. Additional fine tuning includes detecting region edges and merging, and morphological operators to eliminate noise from the spots. the results obtained oil various microarray images show that the proposed technique is quite promising for segmentation of DNA microarray images, obtaining a very high accuracy on background and noise separation.
the goal of this study was to examine detection thresholds of three Chinese vowels in longterm speech-shaped (LTSS) noise. Results showed that vowel detection thresholds were significantly affected by Mandarin tones, ...
详细信息
ISBN:
(纸本)9781424447138
the goal of this study was to examine detection thresholds of three Chinese vowels in longterm speech-shaped (LTSS) noise. Results showed that vowel detection thresholds were significantly affected by Mandarin tones, but not by vowel category. An auditory simulation model based on excitation-pattern computation was conducted to predict detection thresholds in LTSS noise and indicated that the pattern of predicted thresholds was not well matched withthe pattern of behavioral thresholds. these results suggest that detection of Chinese vowels was dependent on not only vowel spectrum, but also tonal pitch, especially when tonal pitch changes dramatically over the time.
A low cost, fast and convenient method based on electronic nose for determination of ammonia nitrogen in wastewater is developed in this study. the pH of the sample is adjusted to higher than 11 by adding saturated so...
详细信息
ISBN:
(纸本)9781424447138
A low cost, fast and convenient method based on electronic nose for determination of ammonia nitrogen in wastewater is developed in this study. the pH of the sample is adjusted to higher than 11 by adding saturated sodium hydroxide solution, through which most of the ammonium ions transform into ammonia and volatilize to the gas phase following Henry's law. A sensor array consist of four metal oxide semiconductor gas sensors responses to the headspace gas of the sample. Response data is analyzed using BP neural network. the result shows a robust relationship between the responses patterns of the electronic nose and the wastewater ammonia nitrogen concentrations.
Accurately predicting the endpoints of chemical compounds is an important step towards drug design and molecular screening in particular. Here we develop a recursive architecture that is capable of mapping Undirected ...
详细信息
ISBN:
(纸本)9783642040306
Accurately predicting the endpoints of chemical compounds is an important step towards drug design and molecular screening in particular. Here we develop a recursive architecture that is capable of mapping Undirected Graphs into individual labels, and apply it to the prediction of a number of different properties of small molecules. the results we obtain are generally state-of-the-art. the final model is completely general and may be applied not only to prediction of molecular properties, but to a vast range of problems in which the input is a graph and the output is either a single property or (with small modifications) a set of properties of the nodes.
this study approaches the problem for damage diagnosis and defect localization in structures applying a patternrecognition procedure combined with substructuring. the problems for damage detection and localization ar...
详细信息
this study approaches the problem for damage diagnosis and defect localization in structures applying a patternrecognition procedure combined with substructuring. the problems for damage detection and localization are first defined as patternrecognition (PR) problems for differentiating between two possible categories of damaged and non-damaged structures (substructures). the problem for feature extraction from frequency response functions is discussed. A stochastic PR procedure is introduced in the context of defect detection and localization problem. the method is demonstrated on a test case of a cantilevered beam. It shows rather accurate performance and low error with simulated and noise contaminated data.
this book constitutes the refereed proceedings of the 4thinternationalconference on patternrecognition and Machine Intelligence, PReMI 2011, held in Moscow, Russia in June/July 2011. the 65 revised papers presented...
ISBN:
(数字)9783642217869
ISBN:
(纸本)9783642217852
this book constitutes the refereed proceedings of the 4thinternationalconference on patternrecognition and Machine Intelligence, PReMI 2011, held in Moscow, Russia in June/July 2011. the 65 revised papers presented together with 5 invited talks were carefully reviewed and selected from 140 submissions. the papers are organized in topical sections on patternrecognition and machine learning; image analysis; image and video information retrieval; natural language processing and text and data mining; watermarking, steganography and biometrics; soft computing and applications; clustering and network analysis; bio and chemo analysis; and document image processing.
A new functional model for burst firing in the dorsal thalamus is proposed where thalamocortical patternrecognition systems, based on kernel machine principles, are connected by burst signaling. the systems include i...
详细信息
ISBN:
(纸本)9781424420728
A new functional model for burst firing in the dorsal thalamus is proposed where thalamocortical patternrecognition systems, based on kernel machine principles, are connected by burst signaling. the systems include input trapping in the dorsal thalamus, cortical learning state memory and processing in the thalamic reticular nucleus. Misclassified events are captured as training examples in the waking state and the patternrecognition systems are trained by extensive thalamic bursting in deep sleep.
Clustering is a, widely used unsupervised data analysis technique in machine learning. However, a common requirement amongst many existing clustering methods is that all pairwise distances between patterns must be com...
详细信息
ISBN:
(纸本)9783642040306
Clustering is a, widely used unsupervised data analysis technique in machine learning. However, a common requirement amongst many existing clustering methods is that all pairwise distances between patterns must be computed in advance. this makes it computationallly expensive and difficult to cope with large scale data used in several applications, such as in bioinformatics. In this paper we propose a novel sequential hierarchical clustering technique that initially builds a hierarchical tree from a small fraction of the entire data, while the remaining data is processed sequentially and the tree adapted constructively. Preliminary results using this approach show that the quality of the clusters obtained does not degrade while reducing the computational needs.
the possibility of application of the knowledge base technology to solve the patternrecognition problems has been considered. For this purpose. the investigations to deepen the pilotage information analysing and proc...
详细信息
the possibility of application of the knowledge base technology to solve the patternrecognition problems has been considered. For this purpose. the investigations to deepen the pilotage information analysing and processing have been carried out and the tasks related to pilotage technique controlling and assessment have been singled out. Prospects of the application of the fuzzy set theory methods and the decision making using the knowledge about problem domain is demonstrated. (C) 2000 Elsevier Science Ltd. All rights reserved.
To enhance recognition accuracy of isolated words identification with small samples in lipreading, SVM is first introduced to act as classifier in this paper. As SVM is based on structural risk minimization, it solves...
详细信息
ISBN:
(纸本)9781424447138
To enhance recognition accuracy of isolated words identification with small samples in lipreading, SVM is first introduced to act as classifier in this paper. As SVM is based on structural risk minimization, it solves the problem of patternrecognition under small samples, on the other hand, it avoids the unreasonable hypothesis in traditional classifier. To meet the requirement of fixed input feature dimensionality in SVM, several input feature dimensionality normalization methods were discussed and compared, including 3-4-3 data segmenting method, HMM based method and DTAK(Dynamic Time Alignment Kernel) based method. Two experiments were performed on the bimodal database, In the first experiment different input feature normalization algorithm were compared on SVM. Experiments showed that DTAK based normalization achieved the best result, in the second experiments SVM was compared with HMM under different number samples occasion. Experimental results showed that SVM performs better than HMM under small samples.
暂无评论