基于图像分析的后视镜失真率自动检测方法

传统的机动车后视镜失真率检测方法采用模拟相机拍照、照片冲洗、人工测量等步骤,效率低下且人为因素对测量结果影响较大。本文使用数字相机成像,针对数字图像,设计并实现了基于图像分析的失真率自动检测算法,通过圆心检测、图像分割、直线与圆交点检测等几个步骤,完成了自动检测功能。该方法高效、准确,达到了实用水平。     

Localized Model and Arithmetic System Based on Two Image Sensors Under Complex Circumstance

Two image sensors simulate directly the way of disposing images with the human’s two eyes, so it has important value to apply in many domains, such as object identification, small unmaned aerial vehicle (UAV), workpiece localization, robot navigation and so on. The object localization based on two image sensors is studied in this paper. It concentrates on how to apply two charge coupled device (CCD) image sensors to object localization of sphere in complex environments. At first a space model of the two ima...     

一种用于红外小目标检测的改进Canny算法

针对红外小目标检测问题,分析了Canny算法存在的不足,将子波理论引入到Canny算法中,提出了基于子波去噪的改进Canny算法,最后用实例证明该算法具有良好的红外小目标检测性能。     

Robust Interactive Method for Hand Gestures Recognition Using Machine Learning

The Hand Gestures Recognition (HGR) System can be employed to facilitate communication between humans and computers instead of using special input and output devices. These devices may complicate communication with computers especially for people with disabilities. Hand gestures can be defined as a natural human-to-human communication method, which also can be used in human-computer interaction. Many researchers developed various techniques and methods that aimed to understand and recognize specific hand gestures by employing one or two machine learning algorithms with a reasonable accuracy. This work aims to develop a powerful hand gesture recognition model with a 100% recognition rate. We proposed an ensemble classification model that combines the most powerful machine learning classifiers to obtain diversity and improve accuracy. The majority voting method was used to aggregate accuracies produced by each classifier and get the final classification result. Our model was trained using a self-constructed dataset containing 1600 images of ten different hand gestures. The employing of canny's edge detector and histogram of oriented gradient method was a great combination with the ensemble classifier and the recognition rate. The experimental results had shown the robustness of our proposed model. Logistic Regression and Support Vector Machine have achieved 100% accuracy. The developed model was validated using two public datasets, and the findings have proved that our model outperformed other compared studies.     

A Novel Non-Invasive Method for Extraction of Geometrical and Texture Features of Wood

Non invasive feature detection in wood based application requires exact discrimination between geometrical edges and texture. It has been found that traditional edge detection algorithms are highly sensitive to noise and texture and produces inferior results with wood. The present work encompasses a micro level reconstruction of Palash and Rosewood by using micro X-rays CT scanner. It also encompasses a new edge detection algorithm using newly constructed Chebyshev polynomial based fractional order differentiator. Transform based method has been used for reconstruction purpose. Newly designed fractional order filter has been applied on these reconstructed images. Chebyshev polynomial based fractional order differentiator has been used for filtering operation. Quadrature Mirror Filter (QMF) concept has been used for design of high pass filter and low pass filter. Preprocessing has been performed by using this filter. Canny edge detection algorithm has been used on this preprocessed image. The algorithm has been tested on two different test cases of wood images a) Palash and b) Rosewood. The effect of relaxation coefficient has also been presented and discussed.     

An application to detect the edge and texture of the flower by canny algorithm

For the edge and texture characteristics of the flower, a new improved canny algorithm is proposed. In the method, canny algorithm detects the edge and the texture of the flower perfectly, the image noise is eliminated simultaneously. The method is applied to all bloom stages. The growing status of the flower could be better analyzed.     

虹膜图像定位方法的研究

针对现有虹膜定位算法的局限性,提出了一种新的虹膜定位的方法。对虹膜图像进行预处理,用阈值方法和投影方法简单地检测得到虹膜内边缘;通过线平均的算法对内边缘进行精确定位。外边缘的定位采用改进的canny算子和新的hough变换相结合的方法。实验结果表明,此方法能比较快速准确地定位出虹膜的内外边缘。     

一种改进的虹膜定位方法

虹膜定位是虹膜识别中至关重要的环节,根据实际应用中瞳孔内部会由于反光而形成亮点的情况，本文对粗定位方法进行改进,利用虹膜图像的灰度直方图，首先根据瞳孔的灰度分布特点求二值化闪值，采用圆内弦中点平均方法粗定位其圆心和半径，从而减少精定位的搜索范围和计算量，然后利用圆检测算法在二值化图像中进行内圃精定位,与传统方法相比,能够减少内圃定位于瞳孔内部的错误,实验结果验证了该方法的可行性和有效性．     

一种新的虹膜定位方法

针对现有虹膜定位算法的局限性,提出了一种新的虹膜定位的方法.先对虹膜图像进行预处理,用canny边缘检测算子检测得到虹膜内边缘;再通过二次canny边缘检测对虹膜外边缘进行粗定位,然后分小区域对虹膜外边缘实现多阈值边缘检测,从而获得真实边缘,再利用虹膜内外边缘之间的耦合关系,结合最小二乘法实现虹膜外边缘的定位.实验结果表明,此方法能比较快速准确地定位出虹膜的内外边缘.     

一种基于最小交叉熵的canny边缘检测算法

考虑传统Canny算子在边缘检测中的不足,提出一种融入最小交叉熵的Canny边缘检测算法.传统Canny算子的高低阈值一般人为地设定固定值,当一幅图像的灰度级集中在某一区域时,容易造成虚假边缘;利用最小交叉熵计算图像的高低阚值,可以得到理想的阈值;利用图像灰度的均值和方差计算Canny算子的高斯滤波参数.实验结果表明,算法较好地提取了图像的边缘信息,抑制噪声能力较强,有效地提高了边缘检测的鲁棒性.