Object detection by global contour shape

We present a method for object class detection in images based on global shape. A distance measure for elastic shape matching is derived, which is invariant to scale and rotation, and robust against non-parametric deformations. Starting from an over-segmentation of the image, the space of potential object boundaries is explored to find boundaries, which have high similarity with the shape template of the object class to be detected. An extensive experimental evaluation is presented. The approach achieves a remarkable detection rate of 83-91% at 0.2 false positives per image on three challenging data sets.     

1D correlation filter based class-dependence feature analysis for face recognition

In this paper, a novel one-dimensional correlation filter based class-dependence feature analysis (1D-CFA) method is presented for robust face recognition. Compared with original CFA that works in the two dimensional (2D) image space, 1D-CFA encodes the image data as vectors. In 1D-CFA, a new correlation filter called optimal extra-class origin output tradeoff filter (OEOTF), which is designed in the low-dimensional principal component analysis (PCA) subspace, is proposed for effective feature extraction. Experimental results on benchmark face databases, such as FERET, AR, and FRGC, show that OEOTF based 1D-CFA consistently outperforms other state-of-the-art face recognition methods. This demonstrates the effectiveness and robustness of the novel method.     

基于小波多分辨率分析的 PDF417 定位算法

为提高二维条码自动检测定位的效率, 提出了一种利用小波多分辨率分析进行 PDF417 二维条码自动检测定位的新方法, 该方法在高频子图基于特征向量利用纹理相似性原理, 采用区域增长法产生二维条码数据区域的候选子区域集合, 然后在低频子图利用二维条码的起始符、终止符及其边界特征对候选子区域进行验证, 最后得到二维条码在图像中的位置. 实验证明该方法能准确定位受到不同污损程度的二维条码图像, 特别在污损程度高的情况下, 该方法具有独特的优势.     

基于特征可分性的核函数自适应构造

核函数的选择与构造是支撑向量机研究中的关键问题和难点.该文针对该问题,首先讨论了特征空间的线性可分性,推导了其判别条件.然后,根据特征完全可分条件,基于函数逼近论和核函数的基本性质,提出了自适应多项式核函数和B-样条核函数模型,给出了模型参数的估计算法.实测数据仿真实验结果表明,与经典的核函数相比,该文提出的算法在分类性能上取得了明显改善.     

Signal‐to‐noise ratios for measuring saliency of features extracted by eigenvector methods from ophthalmic arterial Doppler signals

Abstract: Features are used to represent patterns with minimal loss of important information. The feature vector, which is composed of the set of all features used to describe a pattern, is a reduced‐dimensional representation of that pattern. Medical diagnostic accuracies can be improved when the pattern is simplified through representation by important features. By identifying a set of salient features, the noise in a classification model can be reduced, resulting in more accurate classification. In this study, a signal‐to‐noise ratio saliency measure was employed to determine the saliency of input features of recurrent neural networks (RNNs) used in classification of ophthalmic arterial Doppler signals. Eigenvector methods were used to extract features representing the ophthalmic arterial Doppler signals. The RNNs used in the ophthalmic arterial Doppler signal classification were trained for the signal‐to‐noise ratio screening method. The application results of the signal‐to‐noise ratio screening method to the ophthalmic arterial Doppler signals demonstrated that classification accuracies of RNNs with salient input features are higher than those of RNNs with salient and non‐salient input features.     

Feature extraction and dimensionality reduction by genetic programming based on the Fisher criterion

Abstract: Feature extraction helps to maximize the useful information within a feature vector, by reducing the dimensionality and making the classification effective and simple. In this paper, a novel feature extraction method is proposed: genetic programming (GP) is used to discover features, while the Fisher criterion is employed to assign fitness values. This produces non‐linear features for both two‐class and multiclass recognition, reflecting the discriminating information between classes. Compared with other GP‐based methods which need to generate c discriminant functions for solving c‐class (c>2) pattern recognition problems, only one single feature, obtained by a single GP run, appears to be highly satisfactory in this approach. The proposed method is experimentally compared with some non‐linear feature extraction methods, such as kernel generalized discriminant analysis and kernel principal component analysis. Results demonstrate the capability of the proposed approach to transform information from the high‐dimensional feature space into a single‐dimensional space by automatically discovering the relationships between data, producing improved performance.     

Motif patterns in 2D

Motif patterns consisting of sequences of intermixed solid and don’t-care characters have been introduced and studied in connection with pattern discovery problems of computational biology and other domains. In order to alleviate the exponential growth of such motifs, notions of maximal saturation and irredundancy have been formulated, whereby more or less compact subsets of the set of all motifs can be extracted, that are capable of expressing all others by suitable combinations. In this paper, we introduce the notion of maximal irredundant motifs in a two-dimensional array and develop initial properties and a combinatorial argument that poses a linear bound on the total number of such motifs. The remainder of the paper presents approaches to the discovery of irredundant motifs both by offline and incremental algorithms.     

Fault tolerant multiple event detection in a wireless sensor network

With an increasing acceptance of Wireless Sensor Networks (WSNs), the health of individual sensor is becoming critical in identifying important events in the region of interest. One of the key challenges in detecting event in a WSN is how to detect it accurately transmitting minimum information providing sufficient details about the event. At the same time, it is also important to devise a strategy to handle multiple events occurring simultaneously. In this paper, we propose a Polynomial-based scheme that addresses these problems of Event Region Detection (PERD) by having a aggregation tree of sensor nodes. We employ a data aggregation scheme, TREG (proposed in our earlier work) to perform function approximation of the event using a multivariate polynomial regression. Only coefficients of the polynomial (P$P$) are passed instead of aggregated data. PERD includes two components: event recognition and event report with boundary detection. This can be performed for multiple simultaneously occurring events. We also identify faulty sensor(s) using the aggregation tree. Performing further mathematical operations on the calculated P$P$ can identify the maximum (max) and minimum (min) values of the sensed attribute and their locations. Therefore, if any sensor reports a data value outside the [min, max] range, it can be identified as a faulty sensor. Since PERD is implemented over a polynomial tree on a WSN in a distributed manner, it is easily scalable and computation overhead is marginal. Results reveal that event(s) can be detected by PERD with error in detection remaining almost constant achieving a percentage error within a threshold of 10%$10\%$ with increase in communication range. Results also show that a faulty sensor can be detected with an average accuracy of 94%$94\%$ and it increases with increase in node density.     

A two-stage hardware scheduler combining greedy and optimal scheduling

Greedy scheduling heuristics provide a low complexity and scalable albeit particularly sub-optimal strategy for hardware-based crossbar schedulers. In contrast, the maximum matching algorithm for Bipartite graphs can be used to provide optimal scheduling for crossbar-based interconnection networks with a significant complexity and scalability cost. In this paper, we show how maximum matching can be reformulated in terms of Boolean operations rather than the more traditional formulations. By leveraging the inherent parallelism available in custom hardware design, we reformulate maximum matching in terms of Boolean operations rather than matrix computations and introduce three maximum matching implementations in hardware. Specifically, we examine a Pure Logic Scheduler with three dimensions of parallelism, a Matrix Scheduler with two dimensions of parallelism and a Vector Scheduler with one dimension of parallelism. These designs reduce the algorithmic complexity for an N×N$N\times N$ network from O(N³)$O\left(N^3\right)$ to O(1)$O\left(1\right)$, O(K)$O\left(K\right)$, and O(KN)$O\left(KN\right)$, respectively, where K$K$ is the number of optimization steps. While an optimal scheduling algorithm requires K=2N−1$K=2N-1$ steps, by starting with our hardware-based greedy strategy to generate an initial schedule, our simulation results show that the maximum matching scheduler can achieve 99% of the optimal schedule when K=9$K=9$. We examine hardware and time complexity of these architectures for crossbar sizes of up to N=1024$N=1024$. Using FPGA synthesis results, we show that a greedy schedule for crossbars, ranging from 8×8 to 256×256, can be optimized in less than 20 ns per optimization step. For crossbars reaching 1024×1024 the scheduling can be completed in approximately 10 μs with current technology and could reach under 90 ns with future technologies.     

Region-Based Hierarchical Image Matching

This paper presents an approach to region-based hierarchical image matching, where, given two images, the goal is to identify the largest part in image 1 and its match in image 2 having the maximum similarity measure defined in terms of geometric and photometric properties of regions (e.g., area, boundary shape, and color), as well as region topology (e.g., recursive embedding of regions). To this end, each image is represented by a tree of recursively embedded regions, obtained by a multiscale segmentation algorithm. This allows us to pose image matching as the tree matching problem. To overcome imaging noise, one-to-one, many-to-one, and many-to-many node correspondences are allowed. The trees are first augmented with new nodes generated by merging adjacent sibling nodes, which produces directed acyclic graphs (DAGs). Then, transitive closures of the DAGs are constructed, and the tree matching problem reformulated as finding a bijection between the two transitive closures on DAGs, while preserving the connectivity and ancestor-descendant relationships of the original trees. The proposed approach is validated on real images showing similar objects, captured under different types of noise, including differences in lighting conditions, scales, or viewpoints, amidst limited occlusion and clutter.