image sharpness is an important aspect of image quality. It used to measure the degree of focus at the time of image acquisition. It also play an important role for video compression. Here, one new sharpness algorithm...
详细信息
ISBN:
(纸本)9781509037100
image sharpness is an important aspect of image quality. It used to measure the degree of focus at the time of image acquisition. It also play an important role for video compression. Here, one new sharpness algorithm based on gradient shape is introduced in this paper. which is used for no-reference image. The algorithm gets region of Interest in image firstly, then search the edge which can present sharpness information in selected region. It calculates edge transition zone width, then gets gray contrast in edge region, finally a probability summation algorithm model be set by these factors. This algorithm can calculate the sharpness degree of different images. A lot of experimental results show that this sharpness algorithm is effective and keep consistency with human subjective judgment. It can be used to describe the no-reference image sharpness effectively.
Superpixel segmentation refers to represent an image by small regions composed of pixels with the similar characteristics, which can carry more perceptual and semantic meaning than their simple pixel grid counterparts...
详细信息
ISBN:
(纸本)9781509037100
Superpixel segmentation refers to represent an image by small regions composed of pixels with the similar characteristics, which can carry more perceptual and semantic meaning than their simple pixel grid counterparts. Therefore, it is very important to distribute superpixels with different size over an image for describing image details. In this paper, a new superpixel generation method based on an error-controlled strategy is put forward. At first, we use the SLIC method as a pretreatment step to obtain a preliminary segmentation image. After that, we set two thresholds to iteratively perform the region merging and mean pooling operations. One of the thresholds is used as the condition of merging in two superpixels. The other is used to control the convergence of the whole image error. According to the experimental results, our algorithm can fulfill a comparable trade-off between the number of superpixels and image details representation. In addition to the effective distribution of superpixels, our method can keep the boundary of the main object in the image.
images captured in foggy weather often suffer from bad visibility, because of bad atmospheric visibility. The existence of particles, dust and water impurities always fade the color and lead to a bad contrast of the o...
详细信息
ISBN:
(纸本)9781509037100
images captured in foggy weather often suffer from bad visibility, because of bad atmospheric visibility. The existence of particles, dust and water impurities always fade the color and lead to a bad contrast of the observed objects. And then, we will feel indistinct by our eyes. We introduce a novel algorithm of single image that based on hybrid filter in this paper. The algorithm is based on an advanced median filter, and combined with average filter. The visibility of photos has been greatly improved by use of iteration. Experimental results on a variety of haze images demonstrate the effectiveness of the advanced method, which contains the fast speed of Median filter and also gets more distinct in details compared with other algorithms. The main advantage of the proposed algorithm compared with others is its speed. This speed allows visibility restoration to be applied for the first time within real-time processing applications such as sign, lane-marking and obstacle detection.
Automatic segmentation and early diagnosis of brain tumor is a challenging problem in computer vision and it can provide possibility for pre-operative planning, and solve the problem such as low accurateness and time-...
详细信息
ISBN:
(纸本)9781509037100
Automatic segmentation and early diagnosis of brain tumor is a challenging problem in computer vision and it can provide possibility for pre-operative planning, and solve the problem such as low accurateness and time-consuming in traditional manual segmentation. Under the mentioned problems above, this paper put forward a new method: Based on traditional convolutional neural networks (CNNs), a new architecture model is proposed for automatic brain tumor segmentation, which combines multi-modality images. The newly designed CNNs model automatically learns useful features from multi-modality images to combine multi-modality information. Experiment results show that the proposed model is more accurate than traditional methods and can provide reliable information for clinic treatments.
X-ray phase contrast imaging ( XPCI) shows more detail and capable to produce high contrast images compared to conventional absorb contrast imaging. Not as free-space-propagation-based XPCI which needs a highly cohere...
详细信息
ISBN:
(纸本)9781509037100
X-ray phase contrast imaging ( XPCI) shows more detail and capable to produce high contrast images compared to conventional absorb contrast imaging. Not as free-space-propagation-based XPCI which needs a highly coherent and monochromatic light source such as synchrotron radiation source, several different XPCI methods are practical using desktop X-ray tube, such as micro-focus PCI, grating-based differential PCI and coded aperture-based edge illumination PCI, etc. In this paper, an experimental research on micro-focus XPCI is done with a conventional X-ray source. With a micro collimating aperture, a large focal spot X-ray tube is modified to be a micro focus light source. A fly and feather are used as imaging samples. Sample images are acquired with different energy X-ray such as 20KeV, 40KeV and 60KeV etc. Phase contrast images are compared with absorption contrast images using the same x-ray source setting. Results showed the sharpness caused by phase contrast, and more detail information of the edges inside the imaging samples. The resolution of micro-focus XPCI image reaches to about 10 micro-meters which is not easy to achieve in traditional absorption method.
For about 1800 years, tongue inspection has been one of the four major diagnostic methods in Traditional Chinese Medicine (TCM). The tongue is believed to be able to reflect the health status of the human body. Howeve...
详细信息
ISBN:
(纸本)9781509037100
For about 1800 years, tongue inspection has been one of the four major diagnostic methods in Traditional Chinese Medicine (TCM). The tongue is believed to be able to reflect the health status of the human body. However, making an accurate diagnose with the tongue is not a trivial task. It usually requires enormous training on the TCM doctor before he can make a reasonable diagnosis. Recently, imageprocessing methods have been proposed to automatically process the tongue images and make diagnosis. This study proposes a k-means clustering and adaptive active contour model based automatic tongue region segmentation algorithm. This study is the first step towards the automatic tongue diagnosis. The method was applied on a set of real tongue images. To quantitatively evaluate the segmentation results, the automatically extracted boundaries were compared to the tongue boundaries drawn by experts. An average coverage ratio of 92% was found, indicating the accuracy of the proposed algorithm.
for improving the present situation in the quality detection of plate making product line, a simplified homocentric square filter (SHSF) is proposed and implemented based on machine vision. Firstly, a hardware platfor...
详细信息
ISBN:
(纸本)9781509037100
for improving the present situation in the quality detection of plate making product line, a simplified homocentric square filter (SHSF) is proposed and implemented based on machine vision. Firstly, a hardware platform including camera, lens, light source, and encoder is designed to acquire high quality images. In order to automatically detect defects from the acquired images, the SHSF is designed according to the characteristics of the defects and their background. Furthermore, multiscale analysis is applied to detect defects of different sizes. Combining SHSF with multiscale analysis, nine kinds of defects are detected successfully. Finally, integral image technique (IIT) is applied to the filter for real-time processing. Experimental results show that the proposed method achieves a good performance in accuracy and processing speed of defect detection.
Volumetric parameterization is a basic and important problem for digital geometry processing and computer graphics. This paper presents a volumetric parameterization approach which is based on bounded-distortion harmo...
详细信息
This paper investigates image encryption by using magic squares. We propose a new magic square encryption method to which we refer as Good Lattice Point (GLP) method. Traditional magic square encryption method ([4]) f...
详细信息
ISBN:
(纸本)9781509037100
This paper investigates image encryption by using magic squares. We propose a new magic square encryption method to which we refer as Good Lattice Point (GLP) method. Traditional magic square encryption method ([4]) focuses on disordering positions of image pixel points according to magic squares. GLP method works in the same way but gives different encryption periods and better encrypted effect than the traditional one. In addition, we also propose other new encryption methods which focus on changing values of image pixel points according to operations of magic squares such as addition and production operations. Furthermore, we develop a criteria which measures the effect of image encryption and consequently provides a way to compare different magic square encryption methods.
To promote the forecasting performance of Fuzzy time-series models, based on complex network, a novel fuzzy time series model for stock price forecasting was pressented, this pressented model includes the concept of t...
详细信息
暂无评论