A novel context sensitive multilevel thresholding for image segmentation

Most of the traditional histogram-based thresholding techniques are effective for bi-level thresholding and unable to consider spatial contextual information of the image for selecting optimal threshold. In this article a novel thresholding technique is presented by proposing an energy function to generate the energy curve of an image by taking into an account the spatial contextual information of the image. The behavior of this energy curve is very much similar to the histogram of the image. To incorporate spatial contextual information of the image for threshold selection process, this energy curve is used as an input of our technique instead of histogram. Moreover, to mitigate multilevel thresholding problem the properties of genetic algorithm are exploited. The proposed algorithm is evaluated on the number of different types of images using a validity measure. The results of the proposed technique are compared with those obtained by using histogram of the image and also with an existing genetic algorithm based context sensitive technique. The comparisons confirmed the effectiveness of the proposed technique.     

计算机视觉技术在零件尺寸测量中的应用

介绍了一种应用计算机视觉技术检测机械零件参数的测量方法。以面阵CCD为图像传感器,通过图像采集卡将机械零件的二维图像输入到计算机中。在对原始输入图像进行直方图校正和边缘保持滤波处理后,对得到的较为平滑的零件图像进行边缘检测。利用图像边缘灰度突变的特性,提出了一种结合梯度算子的快速边缘检测方法。并据此计算出零件的各参数值。此种测量方法非常适合于微小、易形变等接触测量难以准确测量的机械零件的参数检测,具有广阔的应用前景。     

一种自适应红外图像增强新算法

针对红外图像经直方图均衡后因部分灰度级被合并而导致细节丢失的问题,提出了一种新的红外图像增强算法——双阈值区分增强算法.该算法首先在原始图像直方图中剔除截断阈值以下的冗余灰度级,接着根据重建直方图的包络确定分界阈值用以区分背景和目标,最后对背景进行直方图均衡处理,对目标进行局部对比度增强及线性灰度变换处理.算法定义了一个自适应参数来适当地抑制背景和噪声,并放大目标细节.相对于传统的直方图均衡算法,所提算法能在增强整体对比度的同时保留图像细节,并且计算量小、实时性好.     

基于区域直方图的图象数据库检索

文章提出了图象区域分割新方法,将图象空间信息与颜色信息有机结合。该方法可以以较少的数据描述一幅图象,有利于图象数据库的管理与检索;解决了尺度变化、旋转和平移不变性问题。     

压缩视频基于内容检索系统的实现

文章介绍了以直方图为工具实现压缩视频基于内容检索系统的方法。其中包括:压缩视频DC图象的获得、DC图象的HSV颜色直方图表示、直方图实现镜头边缘检测、直方图实现镜头代表帧抽取以及代表帧的特征提取等。同时介绍了系统向用户提供的快速浏览功能和查询功能的实现。     

Analog and digital image enhancement using fuzzy expert systems

 Image enhancement is a field that is being used in various areas and disciplines. Advances in computers, microcontrollers and DSP boards have opened new horizons to digital image processing, and have opened many avenues to the design and implementation of new innovative techniques. This paper compares image enhancement via the modification of the probability density function of the gray levels with the new techniques that involves the use of knowledge-base (fuzzy expert) systems that are capable of mimicking the behavior of a human expert. A fuzzy expert system based software for image enhancement, called SmartPhotoLab has been introduced for the above purpose. Present address: A. El-Osery Dept. of Electrical Engineering, New Mexico Tech, Workman Center Rm. 247 801 Leroy place, Socorro, NM 87801 e-mail: elosery@ee.nmt.edu. This work was supported in parts by NASA grants no. NAG2–1196 and 2-1480.     

A high-payload,reversible data hiding scheme based on histogram modification in JPEG bitstream

In this paper, we have presented a high-payload, reversible data hiding scheme that is focused on embedding secret data into joint photographic experts group (JPEG) bitstream by using histogram modification. In JPEG, for AC coefficients, 162 different variable length codes (VLCs) are defined and assigned by Huffman coding for all 162 run/size possibilities. After parsing VLC usage in the bitstream, we generated a histogram of run/size. According to a peak point and the nearest zero point of the histogram, the secret data were embedded by modifying the corresponding VLC values slightly. The experimental results and comparing the performance of our scheme with those of other existing schemes demonstrated that the proposed scheme preserved the quality of the image with no distortion and achieved the goal of high embedding capacity.     

基于人机交互的医学超声图像自动化定量分析

实现医学超声图像的自动化定量分析.设计并实现一个基于人机交互的医学超声图像辅助分析系统.首先由医生凭借临床经验在超声图像中划出目标区域,系统生成目标子图像,然后使用直方图均衡和迭代阈值法对目标子图像进行灰度调整和分割处理,最后根据位图标记结果在原图中提取目标轮廓,并自动生成反映目标大小的长、宽等数据.通过采用一定的图像处理技术,成功地实现了对超声图像中目标区域的分割和自动化定量分析.本系统能弥补当前超声图文管理系统在图像分割和图像分析方面的不足,具有较高的实用性.     

Reversible watermarking method based on asymmetric-histogram shifting of prediction errors

This paper tries to provide a new perspective for the research of reversible watermarking based on histogram shifting of prediction errors. Instead of obtaining one prediction error for the current pixel, we calculate multiple prediction errors by designing a multi-prediction scheme. An asymmetric error histogram is then constructed by selecting the suitable one from these errors. Compared with traditional symmetric histogram, the asymmetric error histogram reduces the amount of shifted pixels, thus improving the watermarked image quality. Moreover, a complementary embedding strategy is proposed by combining the maximum and minimum error histograms. As the two error histograms shift in the opposite directions during the embedding, some watermarked pixels will be restored to their original values, thus the image quality is further improved. Experimental findings also show that the proposed method re-creates watermarked images of higher quality that carry larger embedding capacity compared to conventional symmetric histogram methods, such as Tsai et al.’s and Luo et al.’s works.