A novel method to look for the hysteresis thresholds for the Canny edge detector

In the last few years, several works have been proposed to solve the problem of determining the hysteresis thresholds in an unsupervised way. In this paper, a novel method to solve this problem is proposed. Given a set of candidates for hysteresis thresholds, the basic idea of the proposed method is to combine gradient information with information obtained when the linking process is applied to all candidates. Using the same dataset and the same evaluation methodology already proposed by other works, the results obtained by our method show a performance better than that of the previous methods. The results obtained by the proposed method have been validated only for the Canny edge detector, but there are no restrictions on applying the proposed method to any other edge detector whose strategy is based on the hysteresis mechanism.     

Solving the process of hysteresis without determining the optimal thresholds

Hysteresis is an important edge detection technique, but the unsupervised determination of hysteresis thresholds is a difficult problem. Thus, hysteresis has limited practical applicability. Unimodal thresholding techniques are another edge detection method. They are useful, because the histogram of a feature image (usually the feature image is an approximation of the gradient image) is unimodal, and there are many unsupervised methods to solve this problem. But such techniques do not use spatial information to detect edge points, so their performance is worse than that of the hysteresis.In this paper, we show how to formulate the hysteresis process as a unimodal thresholding problem without determining the optimal hysteresis thresholds. Using similar steps of the Canny edge detector to obtain an approximation of the gradient image we compare the performance of our method against that of a method that determines the best parameters of an edge detector and show that our method performs relatively well. Additionally, our method can adjust its sensitivity by using different unimodal thresholding techniques.     

Variable factorization model based on numerical optimization for hyperspectral anomaly detection

The objective of this article is to develop an anomaly detector as an analytical expression for detecting anomalous objects in remote sensing using hyperspectral imaging. Conventional anomaly detectors based on the subspace model have a parameter which is the dimension of the clutter subspace. The range of possible values for this parameter is typically large, resulting in a large number of images of detector output to be analyzed. An anomaly detector with a different parameter is proposed. The pixel of known random variables from a data cube is modeled as a linear transformation of a set of unknown random variables from the clutter subspace plus an error of unknown random variables in which the transformation matrix of constants is also unknown. The dimension of the clutter subspace for each spectral component of the pixel can vary, hence some elements in the transformation matrix are constrained to be zeros. The anomaly detector is the Mahalanobis distance of the resulting residual. The experimental results which are obtained by implementing the anomaly detector as a global anomaly detector in unsupervised mode with background statistics computed from hyperspectral data cubes with wavelengths in the visible and near-infrared range show that the parameter in the anomaly detector has a significantly reduced number of possible values in comparison with conventional anomaly detectors.     

一种新的基于统计向量和神经网络的边缘检测方法

通过构造不同的统计量定量描述了边缘点邻域灰度的分布特征,并将4个统计量组成统计向量.计算训练图像的统计向量作为样本对BP神经网络训练,然后将训练的BP网络直接用于边缘检测.新方法在统计向量的构造上充分考虑了边缘点和噪声点的区别,具有较好的抗噪性能;BP网络的结构和训练都比较简单;而且不需要设定阈值检测边缘.实验表明,新方法抗噪性能好,达到了令人满意的边缘检测效果.     

基于视觉感知的双层次阈值边缘连接方法

边缘是图像目标的重要特征，但通常边缘检测得到的边缘存在不连续现象，为此在分析传统边缘连接方法的基础上，利用人类视觉系统对边缘连接的多层次感知机理，提出了一种基于视觉感知的双层次边缘连接方法。该方法首先利用大、小阈值产生相应的大、小阈值图像计算其差值以确定模糊边缘点;然后利用人类视觉系统对边缘连接的感知特性系数判别模糊边缘点中真正的边缘点，并将真正的边缘点添加到大阈值图像，使大部分重要的区域边缘能够形成完整的封闭轮廓。仿真实验结果表明，该方法能够有效地改善边缘检测后的边缘不连续现象，相比一些传统的边缘连接方法，运算速度较快，连接效果较好，能满足边缘检测的轮廓封闭性要求。     

基于多特征融合的前向车辆检测方法

针对传统车辆检测方法定位精度不高的问题,提出一种基于多特征融合的前向车辆检测方法。采用基于直方图分析和自适应双阈值的方法分别实现阴影和边缘特征的准确分割,并通过阴影和边缘特征的综合分析,生成车辆假设区域。利用对称性、纹理和轮廓匹配度3个特征融合得到的综合特征对获得的车辆假设区域进行验证,剔除其中的误检区域。实验结果证明,该方法能在不同光照条件下自适应地进行车辆检测,检测率可达92%以上,且在检测率和误检率2项指标上均优于传统基于学习的方法。     

Circle detection using discrete differential evolution optimization

This paper introduces a circle detection method based on differential evolution (DE) optimization. Just as circle detection has been lately considered as a fundamental component for many computer vision algorithms, DE has evolved as a successful heuristic method for solving complex optimization problems, still keeping a simple structure and an easy implementation. It has also shown advantageous convergence properties and remarkable robustness. The detection process is considered similar to a combinational optimization problem. The algorithm uses the combination of three edge points as parameters to determine circle candidates in the scene yielding a reduction of the search space. The objective function determines if some circle candidates are actually present in the image. This paper focuses particularly on one DE-based algorithm known as the discrete differential evolution (DDE), which eventually has shown better results than the original DE in particular for solving combinatorial problems. In the DDE, suitable conversion routines are incorporated into the DE, aiming to operate from integer values to real values and then getting integer values back, following the crossover operation. The final algorithm is a fast circle detector that locates circles with sub-pixel accuracy even considering complicated conditions and noisy images. Experimental results on several synthetic and natural images with varying range of complexity validate the efficiency of the proposed technique considering accuracy, speed, and robustness.     

λτ-space representation of images and generalized edgedetector

An image and surface representation based on regularization theory is introduced in this paper. This representation is based on a hybrid model derived from the physical membrane and plate models. The representation, called the λτ-representation, has two dimensions; one dimension represents smoothness or scale while the other represents the continuity of the image or surface. It contains images/surfaces sampled both in scale space and the weighted Sobolev space of continuous functions. Thus, this new representation can be viewed as an extension of the well-known scale space representation. We have experimentally shown that the proposed hybrid model results in improved results compared to the two extreme constituent models, i.e., the membrane and the plate models. Based on this hybrid model, a generalized edge detector (GED) which encompasses most of the well-known edge detectors under a common framework is developed. The existing edge detectors can be obtained from the generalized edge detector by simply specifying the values of two parameters, one of which controls the shape of the filter (τ) and the other controls the scale of the filter (λ). By sweeping the values of these two parameters continuously, one can generate an edge representation in the λτ space, which is very useful for developing a goal-directed edge detection scheme for a specific task. The proposed representation and the edge detector have been evaluated qualitatively and quantitatively on several different types of image data such as intensity, range, and stereo images     

A data mining approach to face detection

In this paper, we propose a novel face detection method based on the MAFIA algorithm. Our proposed method consists of two phases, namely, training and detection. In the training phase, we first apply Sobel's edge detection operator, morphological operator, and thresholding to each training image, and transform it into an edge image. Next, we use the MAFIA algorithm to mine the maximal frequent patterns from those edge images and obtain the positive feature pattern. Similarly, we can obtain the negative feature pattern from the complements of edge images. Based on the feature patterns mined, we construct a face detector to prune non-face candidates. In the detection phase, we apply a sliding window to the testing image in different scales. For each sliding window, if the slide window passes the face detector, it is considered as a human face. The proposed method can automatically find the feature patterns that capture most of facial features. By using the feature patterns to construct a face detector, the proposed method is robust to races, illumination, and facial expressions. The experimental results show that the proposed method has outstanding performance in the MIT-CMU dataset and comparable performance in the BioID dataset in terms of false positive and detection rate.     

Multi-circle detection on images using artificial bee colony (ABC) optimization

Hough transform has been the most common method for circle detection, exhibiting robustness, but adversely demanding considerable computational effort and large memory requirements. Alternative approaches include heuristic methods that employ iterative optimization procedures for detecting multiple circles. Since only one circle can be marked at each optimization cycle, multiple executions ought to be enforced in order to achieve multi-detection. This paper presents an algorithm for automatic detection of multiple circular shapes that considers the overall process as a multi-modal optimization problem. The approach is based on the artificial bee colony (ABC) algorithm, a swarm optimization algorithm inspired by the intelligent foraging behavior of honeybees. Unlike the original ABC algorithm, the proposed approach presents the addition of a memory for discarded solutions. Such memory allows holding important information regarding other local optima, which might have emerged during the optimization process. The detector uses a combination of three non-collinear edge points as parameters to determine circle candidates. A matching function (nectar-amount) determines if such circle candidates (bee-food sources) are actually present in the image. Guided by the values of such matching functions, the set of encoded candidate circles are evolved through the ABC algorithm so that the best candidate (global optimum) can be fitted into an actual circle within the edge-only image. Then, an analysis of the incorporated memory is executed in order to identify potential local optima, i.e., other circles. The proposed method is able to detect single or multiple circles from a digital image through only one optimization pass. Simulation results over several synthetic and natural images, with a varying range of complexity, validate the efficiency of the proposed technique regarding its accuracy, speed, and robustness.     

Edge detection of noisy digital image using optimization of threshold and self organized map neural network

The purpose of this research is to find a suitable method for detecting the edges of noisy digital images by eliminating the noise effects. The image will be partitioned into equal partitions and the initial threshold of that image partition will be calculated. By applying all these thresholds into the self-organized map (SOM) neural network input optimized for learning and training based optimization algorithm (TLBO), threshold clustering will be performed. The partitioned image will be edge detected by entropy method. Choosing the threshold for image segmentation is of great importance. The mean of the brightness of digital noise images is not a good representative of the initial threshold. Noise causes the mean intensity of the brightness to take distance from the main range of the intensity of the image so the resulting edge detected image will be severely noisy and truncated. By determining the highest frequency of brightness intensity instead of the mean brightness, the above-mentioned weaknesses will be eliminated. This method outperforms many current methods, such as Tsallis entropy, Singh and Kiani and even Canny Edge Detection which demonstrates the effectiveness of the proposed method, In the Table 1 the PSNR of image 5 of the proposed method is 61.4896, but Singh method which is 55.61, Tsallis method which is 53.9234, Kiani method which is 53.9315 the proposed method is less than the other methods.

     

一种改进的汉语全文无指导词义消歧方法

针对现存的基于EM (Expectation maximization)迭代的无指导词义消歧方法收敛缓慢、计算量大的问题, 利用互信息和Z-测试结合的方法选取特征, 并通过一种 统计学习算法估算初始参数值. 实验结果表明改进方法有效地提高了汉语词义消歧的准确率, 具有良好的扩展性和实用性.     

Automatic multiple circle detection based on artificial immune systems

Hough transform (HT) has been the most common method for circle detection, exhibiting robustness but adversely demanding a considerable computational load and large storage. Alternative approaches for multiple circle detection include heuristic methods built over iterative optimization procedures which confine the search to only one circle per optimization cycle yielding longer execution times. On the other hand, artificial immune systems (AIS) mimic the behavior of the natural immune system for solving complex optimization problems. The clonal selection algorithm (CSA) is arguably the most widely employed AIS approach. It is an effective search method which optimizes its response according to the relationship between patterns to be identified, i.e. antigens (Ags) and their feasible solutions also known as antibodies (Abs). Although CSA converges to one global optimum, its incorporated CSA-Memory holds valuable information regarding other local minima which have emerged during the optimization process. Accordingly, the detection is considered as a multi-modal optimization problem which supports the detection of multiple circular shapes through only one optimization procedure. The algorithm uses a combination of three non-collinear edge points as parameters to determine circles candidates. A matching function determines if such circle candidates are actually present in the image. Guided by the values of such function, the set of encoded candidate circles are evolved through the CSA so the best candidate (global optimum) can fit into an actual circle within the edge map of the image. Once the optimization process has finished, the CSA-Memory is revisited in order to find other local optima representing potential circle candidates. The overall approach is a fast multiple-circle detector despite considering complicated conditions in the image.     

Robust edge detection

Edge detection is an important issue in computer vision and image understanding systems. Most conventional techniques have assumed Gaussian noise, and their performance could decrease with the departure of noise distribution from normality. In this paper, we present an edge detection approach using robust statistics. The edge structure is first detected by a robust one-way design model, and then localized by a robust contrast test. Finally, hysteresis thresholding is applied to yield the output edge map. To evaluate its performance, experiments were carried out on synthetic and real images corrupted with both Gaussian noise and a mixture of Gaussian and impulsive noise. The results show that the performance of the proposed edge detector is stable and reliable under severe impulsive noise conditions.     

智能无人机轨迹与任务卸载联合优化

移动边缘计算（MEC）是云计算技术在边缘基础设施之上的应用拓展。考虑一个高能效的无人机移动边缘计算系统,通过联合优化无人机的运动轨迹、任务卸载策略和计算资源分配来最小化系统的能耗。为解决以上问题,提出一种双层优化方法,在上层用基于无监督学习的信道增益-自组织特征映射网络（h-SOM）对用户进行实时聚类,该聚类是以信道增益作为判断类别的指标并得到无人机的最佳部署位置;在下层根据无人机的部署,将计算卸载和计算资源分配问题转化为混合整数非线性规划问题（MINLP）,并采用带有精英初始策略和自适应双变异策略的改进差分进化算法（IDE）进行迭代求解,精英初始策略可以根据h-SOM的聚类结果提供优秀的初始解,自适应双变异策略能够提高算法的全局搜索能力并促进算法收敛,从而获得更好的任务卸载决策。通过仿真实验验证了所提方法的有效性,并与传统算法进行了比较,其优化效果显著,为MEC系统的联合优化提供了一种新思路。     

新颖的物体轮廓提取：基于双边滤波的多级边缘检测*

提出了一种新颖的物体轮廓提取方法,即通过多级边缘检测来提取物体的主要边界。双边滤波器用来建立多级,同时Canny边缘算子相应地产生边缘图,组合边缘图构造出一幅多级图。次段被定义为边缘像素的连接结构,提取并连接这些次段可以构成闭合轮廓。最终图像中最相关的闭合轮廓被判定为真实的物体轮廓。实验结果表明,该物体轮廓提取方法具有较高的可靠性并且受噪声影响较小。     

一种有效的空间域视频传输差错掩盖算法

针对易错信道中传输视频图像容易发生差错而导致图像块丢失的现象，提出一种有效的空间域差错掩盖算法．根据周围正确解码块中的边缘信息把丢失块分成平滑块和边缘块两类．对于不舍有边缘的丢失块用简单的线性插值进行恢复；对含有边缘的丢失块先用基于梯度的自适应预测（GAP）法获得丢失块的初始值，然后用最大后验概率方法对初始值进行优化．实验结果表明该算法能够获得优越的图像质量，同时计算复杂度低，适用于视频图像的实时性传输．     

A novel low false alarm rate pedestrian detection framework based on single depth images

Pedestrian detection is an important image understanding problem with many potential applications. There has been little success in creating an algorithm which exhibits a high detection rate while keeping the false alarm in a relatively low rate. This paper presents a method designed to resolve this problem. The proposed method uses the Kinect or any similar type of sensors which facilitate the extraction of a distinct foreground. Then potential regions, which are candidates for the presence of human(s), are detected by employing the widely used Histogram of Oriented Gradients (HOG) technique, which performs well in terms of good detection rates but suffers from significantly high false alarm rates. Our method applies a sequence of operations to eliminate the false alarms produced by the HOG detector based on investigating the fine details of local shape information. Local shape information can be identified by efficient utilization of the edge points which, in this work, are used to formulate the so called Shape Context (SC) model. The proposed detection framework is divided in four sequential stages, with each stage aiming at refining the detection results of the previous stage. In addition, our approach employs a pre-evaluation stage to pre-screen and restrict further detection results. Extensive experimental results on the dataset created by the authors, involves 673 images collected from 11 different scenes, demonstrate that the proposed method eliminates a large percentage of the false alarms produced by the HOG pedestrian detector.     

Orbit determination using incremental phase and TDOA of X-ray pulsar

X-ray pulsars offer stable, periodic X-ray pulse sequences that can be used in spacecraft positioning systems. A method using X-ray pulsars to determine the initial orbit of a satellite is presented in this paper. This method suggests only one detector to be equipped on the satellite and assumes that the detector observes three pulsars in turn. To improve the performance, the use of incremental phase in one observation duration is proposed, and the incremental phase is combined with the time difference of arrival (TDOA). Then, a weighted least squares (WLS) algorithm is formulated to calculate the initial orbit. Numerical simulations are performed to assess the proposed orbit determination method.     

A high performance edge detector based on fuzzy inference rules

Edge detection is an important topic in computer vision and image processing. In this paper, a novel edge detector based on fuzzy If-Then inference rules and edge continuity is proposed. The fuzzy If-Then rule system is designed to model edge continuity criteria. The maximum entropy principle is used in the parameter adjusting process. We also discuss the related issues in designing fuzzy edge detectors. We compare it with the popular edge detectors: Sobel and Canny edge detectors. The proposed fuzzy edge detector does not need parameter setting as Canny edge detector does, and it can preserve an appropriate detection in details. It is very robust to noise and can work well under high level noise situations, while other edge detectors cannot. The detector efficiently extracts edges in images corrupted by noise without requiring the filtering process. The experimental results demonstrate the superiority of the proposed method to existing ones.