Textual and shape-based feature extraction and neuro-fuzzy classifier for nuclear track recognition

Track counting algorithms as one of the fundamental principles of nuclear science have been emphasized in the recent years. Accurate measurement of nuclear tracks on solid-state nuclear track detectors is the aim of track counting systems. Commonly track counting systems comprise a hardware system for the task of imaging and software for analysing the track images. In this paper, a track recognition algorithm based on 12 defined textual and shape-based features and a neuro-fuzzy classifier is proposed. Features are defined so as to discern the tracks from the background and small objects. Then, according to the defined features, tracks are detected using a trained neuro-fuzzy system. Features and the classifier are finally validated via 100 Alpha track images and 40 training samples. It is shown that principle textual and shape-based features concomitantly yield a high rate of track detection compared with the single-feature based methods.     

A novel method for detecting and counting overlapping tracks in SSNTD by image processing techniques

Overlapping object detection and counting is a challenge in image processing. A new method for detecting and counting overlapping circles is presented in this paper. This method is based on pattern recognition and feature extraction using “neighborhood values“ in an object image by implementation of image processing techniques. The junction points are detected by assignment of a value for each pixel in an image. As is shown, the neighborhood values for junction points are larger than the values for other points. This distinction of neighborhood values is the main feature which can be utilized to identify the junction points and to count the overlapping tracks. This method can be used for recognizing and counting charged particle tracks, blood cells and also cancer cells. The method is called “Track Counting based on Neighborhood Values” and is symbolized by “TCNV”.     

Alternative procedure for LR 115 chemical etching and alpha tracks counting

An improved procedure for etching and analysis of alpha tracks induced in LR-115 detectors is proposed with the advantages of simplicity and its relatively low cost. A new type of detector holder was designed to etch and rinse efficiently up to 100 detectors. We develop a simple and reliable methodology with a semiautomatic track count using a Nikon digital camera coupled to a PC and employing software “SCION” freely available on the Internet. Track images are binarized prior the application of software “SCION” so that original track shapes are not distorted, space resolution is improved and track counting has low dependence on focus and illumination level. High discrimination for tracks is achieved when marks and rips perturb the detector surface. An image generator of nuclear tracks is included to study the effect of track overlapping effect on counting.     

Supervised nuclear track detection of CR-39 detectors by cellular automata

In this paper, cellular automata are used to detect the nuclear tracks in the track images captured from the surface of CR-39 detectors. Parameters of the automaton as the states, neighborhood, rules and quality parameters are defined optimally for the track image data set under analysis. The presented method is a supervised computational algorithm which comprises a rule definition phase as the learning procedure. Parameter optimization is also performed to adapt the algorithm to the data set used.     

Statistical variance reduction analysis in counting the highly overlapping nuclear tracks

In this paper, a statistical analysis is performed on counting the highly overlapping tracks. As the track density and tracks overlapping increase, the statistical error of track counting raises. An upper limit for the uncertainty of the track counting is found. The approximated number of tracks is computed based on the measured number of objects, approximated ratio of overlapping cases, measured number of tracks in the overlapping cases and calculated uncertainty. The approximation of the number of tracks obtained through both calculation and measurement processes is validated comparing with the real number of manually counted tracks and the uncertainty is calculated for each comparison.     

Counting recoil proton tracks on PADC without a pre-etching step. A novel approach for neutron dosimeter application

At many laboratories involving in routine, individual neutron dosimetry, poly allyl diglycol carbonate (PADC) is utilized for particle registration. A pre-etching step is commonly applied in order to remove the alpha tracks of the environmental radon, so achieving the required lower limit of detection (LLD) performance. A novel approach is presented, which makes this pre-etching step unnecessary, but ensuring an excellent LLD performance, also providing a good throughput for a routine service. Exposing the PADC material to carbon dioxide atmosphere before the main etching step enhances the visibility of the microscopic track image. The enhancement is so efficient that the track size and image contrast between the alpha and the proton tracks are well distinctive. Customized image analyzer software is able to distinguish alpha and proton tracks, so providing the proton track density separately. A pilot study about the performance and potentials of this novel approach is presented. A complete application working with this approach was introduced to routine neutron dosimeter service 2 years ago and its good performance has been verified.     

像增强器视场缺陷检测方法研究

在数字式像增强器性能综合检测系统的基础上，研究了像增强器视场缺陷的检测方法。通过数字视频摄像机采集缺陷图像，利用数字图像处理方法确定缺陷性质，缺陷处理算法基于图像灰度特征而不是边缘特征提取。该方法适用于对检测对象有具体量化和定位要求，而不是简单的只有数量和形状特征要求的场合。实验表明，该检测方法能够较准确地检测像增强器的缺陷。     

A new method for measuring tracks density in CR-39 detectors by compensating for overlapping tracks

CR-39 nuclear track detectors have been used as an efficient radon monitoring system. Exposing these detectors to radon would result in tracks where ever particles hit the surface of the detector. Natural radioactivity of alpha particles (radon concentration) is measured based on the count of these tracks on CR-39 detector. Counting these tracks by eye is not an easy task especially when the density of the tracks is high. Automated tracks counting systems are attractive solution for this problem. However, as the density of tracks increases the accuracy of these methods deteriorates. This is due to the fact that overlapped tracks are counted as one track during the segmentation process resulting in an underestimation of the real count of the tracks. This work proposes a new image processing based automated approach for tracks counting. This new automated approach solves the problem of counting overlapping tracks by approximating their number and then adjusting the final count of tracks by compensating for those uncounted tracks. The experimental results show that the proposed approach reduces the counting percentage error from 5.64% to 2.54% by solving the counting problem of the overlapped tracks.     

Geometrical event reconstruction in the COMPASS experiment

The procedure of geometrical event reconstruction used at the setup COMPASS (CERN/SPS) including the algorithm of charged particle tracks reconstruction and the algorithm of the search of the vertex of beam muon-target interaction is presented. The results of the calculation of the accuracy of reconstructed objects and efficiency of the track reconstruction algorithm are given.     

Defects’ geometric feature recognition based on infrared image edge detection

Edge detection is an important technology in image segmentation, feature extraction and other digital image processing areas. Boundary contains a wealth of information in the image, so to extract defects’ edges in infrared images effectively enables the identification of defects’ geometric features. This paper analyzed the detection effect of classic edge detection operators, and proposed fuzzy C-means (FCM) clustering-Canny operator algorithm to achieve defects’ edges in the infrared images. Results show that the proposed algorithm has better effect than the classic edge detection operators, which can identify the defects’ geometric feature much more completely and clearly. The defects’ diameters have been calculated based on the image edge detection results.     

一种基于混合人工鱼群的图像模糊增强算法

模糊隶属度函数的形式直接影响灰度图像增强的质量。为进一步改善图像模糊增强的效果,对目前的模糊隶属度函数进行研究,并提出一种改进的参数化 型模糊隶属度函数用于图像增强。所提算法利用图像对比度的质量评价模型,结合人工鱼群算法和Powell算法搜索 型函数中的未知参数值,进而确定该模糊隶属度函数。通过实验结果表明：该算法能够较好的改善灰度图像质量,并且控制参数可通过优化算法自适应获得,具有较好的通用性,是一种有效的图像模糊增强算法。     

被动式人体太赫兹图像中的目标辅助识别

提出一种被动式人体太赫兹安检仪的图像目标标注方法,用于辅助识别被安检人员隐藏携带的违禁物品。分析了探测器中心频率为0.1THz的被动式太赫兹人体安检仪图像特点,根据其特点进行图像增强、分割、轮廓提取等处理,进而实现图像中的目标检测,将其目标位置映射在同场景同时间采集的可见光图像上。实验结果表明,该方法可以有效地提取并标示可疑目标,帮助快速发现衣物下的隐匿物品,提高太赫兹安检仪的实用性。     

被动式人体太赫兹图像中的目标辅助识别

提出一种被动式人体太赫兹安检仪的图像目标标注方法,用于辅助识别被安检人员隐藏携带的违禁物品。分析了探测器中心频率为0.1 THz的被动式太赫兹人体安检仪图像特点,根据其特点进行图像增强、分割、轮廓提取等处理,进而实现图像中的目标检测,将其目标位置映射在同场景同时间采集的可见光图像上。实验结果表明,该方法可以有效地提取并标示可疑目标,帮助快速发现衣物下的隐匿物品,提高太赫兹安检仪的实用性。     

A survey of spectrogram track detection algorithms

The detection of tracks in spectrograms is an important step in remote sensing applications such as the analysis of marine mammal calls and remote sensing data in underwater environments. Recent advances in technology and the abundance of data requires the development of more sensitive detection methods. This problem has attracted researchers’ interest from a variety of backgrounds ranging between image processing, signal processing, simulated annealing and Bayesian filtering. Most of the literature is concentrated in three areas: image processing, neural networks, and statistical models such as the Hidden Markov model. There has not been a review paper which describes and critically analyses the application of these key algorithms. This paper presents an extensive survey and an algorithm taxonomy, additionally each algorithm is reviewed according to a set of criteria relating to their success in application. These criteria are defined to be their ability to cope with noise variation over time, track association, high variability in track shape, closely separated tracks, multiple tracks, the birth/death of tracks, low signal-to-noise ratios, that they have no a priori assumption of track shape and that they are computationally cheap. Our analysis concludes that none of these algorithms fully meets these criteria.     

Infrared image enhancement through saliency feature analysis based on multi-scale decomposition

To improve contrast between dim target region and background in infrared (IR) long-range surveillance, this paper proposes a fast image enhancement approach using saliency feature extraction based on multi-scale decomposition. Firstly, a smooth based multi-scale decomposition is designed and applied to original infrared image, generating sub-images with various frequency components at different decomposition levels. The dim target regions of sub-images are extracted by a local frequency-tuned based saliency feature detection method, secondly. With saliency maps created by saliency extraction using multi-scale local windows with different sizes, the sub-images are enhanced at different decomposition scales. Finally, the enhanced result is reconstructed by synthesizing the all sub-images with adjustable synthetic weights. Since salient areas are analyzed based on fast multi-scale image decomposition, IR image can be s enhanced with good contrast successfully and rapidly. Compared with other algorithms, the experimental results prove that the proposed method is robust and efficient for IR image enhancement.     

基于模糊松弛迭代的分层图像增强算法

映射函数的研究是图像增强技术的关键,模糊松弛技术利用模糊函数建立其映射函数,通过设置该算法的模糊松弛参数可以达到对图像可控增强处理的目的。为了实现针对不同层次的图像内容实现可控式模糊增强,提出了多层次交互迭代的模糊松弛图像增强算法。通过分析各子层图像的直方图,分别设定各子层的模糊松弛参数,并建立对应的模糊松弛映射函数;分别对各子层图像进行可控迭代增强,选择合适的迭代次数作为终止条件。给出的实验结果证明了该算法能够取得较好的增强效果。     

无人机对地目标自动检测与跟踪技术

针对无人机图像帧序列具有平台高速运动,视角旋转强烈,需要实时处理等特点,提出一种基于双级旋转不变特征空间检测（粗匹配-精细匹配）与并行特征提取跟踪的无人机对地目标图像帧序列自动快速目标检测与跟踪算法。采用图像子块的平均灰度值、灰度值方差、灰度值梯度构建特征空间。通过构造图像特征空间的方法来快速筛选待匹配图像的可疑区域,删除大量的背景区域,检测算法使用全局初步匹配加局部精细匹配的方法来规避算法复杂度的缺陷。理论及实验分析表明:该算法实时性强,对图像的旋转畸变具有抵消作用,对异常情况可以恰当处理,且全局初步匹配流程具有可移植性,可以在其他图像匹配跟踪算法中充当预处理器。实验结果表明:该算法在无人机对地的情况下可以保证对地面目标的稳定跟踪,配套检测算法具有较好的实时性,满足无人机图像目标检测跟踪实时处理的需要。     

一种自适应路面图像模糊增强算法

针对传统的图像模糊增强算法增强强度小、处理灰度层次变化丰富的图像效果不佳以及控制参量难以设置等问题,提出了一种新的图像模糊增强算法.首先对局部窗口中像素点进行基于邻域一致性模糊熵测度的分类,并以分类为依据,对每个像素点均确定一最佳渡越点.对模糊隶属度函数也进行了研究改进,设计的函数具有良好的曲线形状,并通过调整控制参量,使渡越点的位置和函数曲线进行最佳的结合,能通过少量的迭代次数获得较好的增强效果.在模糊逆映射上,采用线性逆变换函数,保持了模糊映射所带来的增强效果,并消除了由于截断带来的灰度信息的损失,在运算效率上也得到了提高.新算法对灰度变化丰富的路面图像的增强取得了良好的效果,并且控制参量均为自适应计算,不需进行人为干预,具有很好的通用性.     

Estimate of etched tracks by optical method and spark counting

Solid State Nuclear Track Detectors (SSNTD) are commonly used for long term measurements of radon, thoron and progeny concentrations. In the present study, we compared the optical counting method and the spark counting technique for counting the alpha tracks on LR 115 track detector films. The paper discusses the various parameters that are innate in the process. More than 300 films were counted by both optical microscope and spark counter and the results are compared. The overall results show that the tracks obtained by spark counting are marginally less compared to the optical measurement. A linear fit of the data gives a slope less that one, which indicates that both the methods are almost in good agreement for counting the tracks when the track density is low. At higher track densities the spark counter gives an underestimation with respect to actual number of tracks formed, which could be corrected using a relation obtained between the tracks measured by optical method and spark counting.      

特征增强SSD算法及其在遥感目标检测中的应用

为了提高对复杂场景下多尺度遥感目标的检测精度,提出了基于多尺度单发射击检测(SSD)的特征增强目标检测算法.首先对SSD的金字塔特征层中的浅层网络设计浅层特征增强模块,以提高浅层网络对小目标物体的特征提取能力;然后设计深层特征融合模块,替换SSD金字塔特征层中的深层网络,提高深层网络的特征提取能力;最后将提取的图像特征与不同纵横比的候选框进行匹配以执行不同尺度遥感图像目标检测与定位.在光学遥感图像数据集上的实验结果表明,该算法能够适应不同背景下的遥感目标检测,有效地提高了复杂场景下的遥感目标的检测精度.此外,在拓展实验中,文中算法对图像中的模糊目标的检测效果也优于SSD.