基于CLAHE和开闭运算的绿色苹果图像分割

依据绿色苹果图像的自身特点,设计了一种分区域提取而后合并的图像分割方法。首先对图像进行限制对比度自适应直方图均衡化(contrast limited adaptive histogram equalization,CLAHE),增大果实和背景的颜色差,而后获取R-B色差图像,得到以光常区域为主的区域;然后用CLAHE处理后的图像进行开闭运算,提取局部极大值,得到以高亮区域为主的区域;最后将两区域合并获得完整的果实目标区域。为了验证该方法的有效性,运用Hough算法检测圆,并用相对偏差、圆心相对误差及半径相对误差三个评价指标来定量评价。试验结果表明,提取区域与果实目标区域相比,顺光下,相对偏差、圆心和半径相对误差平均值分别为3.59%,4.76%和2.60%;逆光下,三个指标的平均值分别为10.77%、16.77%、11.49%。无论是顺光还是逆光都有很好的识别效果,能满足机器人采摘果实的精确定位的要求。     

光纤网络中的云数据并行调度

光纤网络的通信质量受到云数据调度均衡性影响很大,为了改善光纤网络通信质量,提高网络中云数据传输的吞吐量和保真率,通过云数据并行调度,实现云数据传输均衡,提出一种基于自适应判决反馈均衡的光纤网络云数据并行调度模型。构建光纤网络通信的传输信道模型,采用最小均方误差估计方法进行光纤网络的量化融合估计,运用匹配滤波检测器进行云数据中的干扰滤波处理,结合自适应判决反馈均衡方法进行信道均衡,在均衡的信道中对滤波输出的云数据进行并行调度和多线程输出调制。仿真结果表明,采用该方法进行光纤网络中的云数据并行调度的均衡性较好,输出数据的保真率较高,误码率较低,改善了光纤网络的整体性能。     

A quantitative measure based infrared image enhancement algorithm using plateau histogram

A quantitative measure based scene-adaptive contrast enhancement algorithm for an infrared (IR) image is proposed. This method regulates the probability density function (PDF) of the raw image firstly, and then applies an improved plateau histogram equalization method whose plateau threshold is determined by the concavity of the regulated PDF to enhance the raw IR image. In the stepped parameter tuning process of the algorithm, quantitative measure EME is used as the criterion to determine the optimal PDF regulator factor and plateau threshold. The above improvements contribute to the performance promotion of the proposed algorithm, whose effectiveness is validated by the final assessment with visual quality and quantitative measures.     

Design of dual-core resonant leaky optical fibre based gain flattening filters for erbium doped fibre amplifiers

We propose optical fibre based filters employing dual-core resonant leaky structure for gain equalization of erbium doped fibre amplifier (EDFA). Spectral loss variation of the structure has been utilized to suppress gain peak and, thus, flatten overall gain profile in the C-band. We show 15.7 dB flat gain with ± 1.6 dB ripple in the wavelength range from 1525 nm to 1555 nm using a single filter and 18±0.7 dB gain using two cascaded filters.     

基于灰度冗余图像融合增强研究

为了改善原图像直方图像素灰度分布集中于低灰度级,与经过常规直方图均衡化处理的图像直方图像素灰度分布集中于高灰度级且均具有灰度冗余的图像增强效果,提出了一种新的图像增强算法.对原始图像进行常规直方图均衡处理,并对原始图像与进行了常规直方图均衡处理的图像分别在尽可能大的范围内进行映射,使两者变换后都尽可能去除灰度冗余而有较大的动态范围.取变换后的两图像各自的优点,融合成一幅效果最好的图像.实验结果验证了理论分析的正确性.     

Multi-level quantization and blind equalization based direct transmission method of digital baseband signal

The direct transmission of digital baseband signals has practical significance in the field of Ethernet terminal connection, high-speed digital communication, data transmission of various types of information peripherals. The signal amplitude gradually decays while the transmission distance increases. Also the attenuation is proportional to the signal frequency, resulting in signal distortion and receiving error. It is a common method for digital baseband signal transmission to use pre-emphasis chip and equalizer chip to improve the transmission quality with a wide range of mature applications. This paper describes a new type of digital signal transmission method, as the receiver using analog-to-digital converter, instead of equalizer chip, to achieve the multi-level quantization of receiving time-domain data waveform. The waveform of the transmitted digital high and low level signal is sampled into multi-bit values. Then, the paper realizes adaptive frequency domain equalization based on soft threshold and makes use of multi-level quantization soft information for error correction. Error correcting code is mainly used to correct the error caused by the channel bandwidth limit, external noise or interference in the process of data transmission, so as to improve the stability and reliability of the transmission. The paper uses the two-stage error correcting codec system based on both Turbo and BCH coding, to achieve the high performance of Turbo code, and good characters of respond time and complexity. The transmitter outputs 12.5 MHz pseudo-random sequence through a 199.93 meter unshielded balanced twisted pair transmission medium. And the receiver circuit using a 62.5MSPS analog-to-digital converter over-samples the waveform to 8-level quantity. The output error of a 65536 bit pseudo-random sequence is less than 8 bits, and the error correction can be further improved by 8b-10b codec. Compared with the traditional pre-emphasis and balanced interface ICs connection, the method described in this article has the advantages of longer transmission distance, better flexibility and wider scope of use.     

图像增强算法综述

图像增强算法能够提高图像整体和局部的对比度,突出图像的细节信息,使增强后的图像更符合人眼的视觉特性且易于机器识别,在军事和民用领域具有广泛的应用。本文从图像增强算法的原理出发,归纳总结了近年来应用比较广泛的4类图像增强算法及其改进算法,包括直方图均衡图像增强算法、小波变换图像增强算法、偏微分方程图像增强算法和基于Retinex理论的图像增强算法。结合人眼视觉特性、噪声抑制、亮度保持和信息熵最大化等图像增强的改进算法,在保证增强图像具有较高对比度的前提下,可进一步提升图像的质量。实现了9种较为典型的图像增强算法,采用主观和客观的评价方法对增强效果进行了对比,分析了不同增强算法的优缺点,并给出了这些算法的计算时间。对这些算法的深入研究能够推动图像增强技术向更高水平发展,从而使图像增强技术在多个学科领域发挥重要作用。     

一种改进的直方图均衡化

从理论上分析了直方图均衡化过程中灰度级合并的原因,给出了简单的物理解释和解析推导,并讨论了该算法的缺点。为了使图像保留更多的细节信息、消除虚假轮廓且保持连续性,提出了保留灰度级的直方图均衡化算法,并分析了其基本原理。实验结果表明:通过该算法处理的图像比采用传统直方图均衡化算法处理的图像效果好。     

基于SOPC设计的红外图像直方图均衡器

基于Altera公司的SOPC-NIOSⅡ嵌入式软核处理器技术,提出了一种实现实时红外图像直方图均衡的设计方法.并在FPGA上实现了这一系统,使用Quartus Ⅱ软件中集成的SOPC Builder工具完成了整个系统的设计.该系统主要包括NIOS Ⅱ主控模块、存储器模块、自定制的直方图统计模块以及用户控制模块.实验结果表明,该系统具有体积小,功耗低,数据处理速度快,可靠性高等特点.     

Molecular electronegativity in density functional theory (VI) --Atom-bond electronegativity equalization model

Ｒｅｃｅｎｔｌｙ,ｄｅｖｅｌｏｐｍｅｎｔａｎｄａｐｐｌｉｃａｔｉｏｎｏｆｄｅｎｓｉｔｙｆｕｎｃｔｉｏｎａｌｔｈｅｏｒｙ(ＤＦＴ)ａｒｅｑｕｉｔｅａｔｔｒａｃｔｉｖｅｔｏｓｃｉｅｎｔｉｓｔｓ’ａｔｔｅｎｔｉｏｎ.Ｔｈｅｒｅｌｅｖａｎｔｐａｐｅｒｓａｒｅｉｎｃｒｅａｓｉｎｇｖｅｒｙｑｕｉｃｋｌｙ[1—11].ＳｉｎｃｅｔｈｅｅｌｅｃｔｒｏｎｅｇａｔｉｖｉｔｙｃｏｎｃｅｐｔｗａｓｐｒｏｐｏｓｅｄｂｙＰａｕｌｉｎｇ[12]ｔｏｄｅｓｃｒｉｂｅｔｈｅｐｏｗｅｒｏｆａｎａｔｏｍｉｎａｍｏｌｅｃｕｌｅｔｏａｔｔｒａｃｔｅｌ…