Improve Fractal Compression Encoding Speed Using Feature Extraction and Self-organization Network

Image compression consists of two main parts: encoding and decoding. One of the important problems of the fractal theory is the long encoding implementation time, which hindered the acceptance of fractal image compression as a practical method. The long encoding time results from the need to perform a large number of domain-range matches, the total encoding time is the product of the number of matches and the time to perform each match. In order to improve encoding speed, a hybrid method combining features extraction and self-organization network has been provided, which is based on the feature extraction approach the comparison pixels by pixels between the feature of range blocks and domains blocks. The efficiency of the new method was been proved by examples.

Effect of compression on hydroscopicity of extracted Chinese fir heartwood

In order to clarify the effects of extraction and compression on the hydroscopicity of wood, Chinese fir （Cunninghamia lanceolata Hook.） heartwood samples with or without extraction were radially or tangentially compressed under water-saturated condition at room temperature. Warm water and 1% sodium hydroxide were used as different solutions for extraction. Water absorption capacity and moisture adsorption isotherms of the compressed samples were then tested. The fractal dimension of internal wood surfaces （Dfs） was calculated based on adsorption isotherms by FHH equation. Results showed that in both compressed groups, the hydroscopicity of samples extracted by sodium hydroxide solution improved greatly, while that of samples extracted by warm water changed little, compared with that of water-saturated samples. Recovery of set and the change of hydroscopic environment inside wood were main reasons for the difference of water absorption among water-saturated samples and samples extracted with warm water and sodium hydroxide solution. The swelling rate of samples extracted by sodium hydroxide solution significantly increased. Moreover, the swelling rate in the tangential direction of tangentially compressed samples was obviously higher than that in radial direction of radially compressed ones. Dfs values of woods extracted by warm water and sodium hydroxide solution decreased by 0.002 and 0.007 in a radially compressed group and by 0.013 and 0.013 in a tangentially compressed group, compared to those of water-saturated one. Therefore, the conclusion can be made that the extraction and compression treatments used in this study have no obvious effects on internal wood surface.

Real-time traffic information extraction based on compressed video with interframe motion vector

Extraction of traffic information from image or video sequence is a hot research topic in intelligenttransportation system and computer vision. A real-time traffic information extraction method based on com-pressed video with interframe motion vectors for speed, density and flow detection, has been proposed for ex-traction of traffic information under fixed camera setting and well-defined environment. The motion vectors arefirst separated from the compressed video streams, and then filtered to eliminate incorrect and noisy vectors u-sing the well-defined environmental knowledge. By applying the projective transform and using the filtered mo-tion vectors, speed can be calculated from motion vector statistics, density can be estimated using the motionvector occupancy, and flow can be detected using the combination of speed and density. The embodiment of aprototype system for sky camera traffic monitoring using the MPEG video has been implemented, and experi-mental results proved the effectiveness of the method proposed.

High Performance of Imaging Extraction for Infrared Satellite Cloud Image

The isotherm is an important feature of infrared satellite cloud images (ISCI), which can directly reveal substantial information of cloud systems. The isotherm extraction of ISCI can remove the redundant information and therefore helps to compress the information of ISCI. In this paper, an isotherm extraction method is presented. The main aggregate of clouds can be segmented based on mathematical morphology. T algorithm and IP algorithm are then applied to extract the isotherms from the main aggregate of clouds. A concrete example for the extraction of isotherm based on IBM SP2 is described. The result shows that this is a high efficient algorithm. It can be used in feature extractions of infrared images for weather forecasts.

Micro-Doppler Parameter Estimation Method Based on Compressed Sensing

A micro-Doppler parameter estimation method based on compressed sensing theory is proposed in this paper.The micro-Doppler parameter estimation algorithm was improved for micro-motion targets with translation in this paper.Relatively ideal micro-Doppler parameter estimation results were obtained.The proposed micro-Doppler parameter estimation was compared with the traditional micro-Doppler parameter estimation algorithm.Requirements for return signal length were analyzed with this new algorithm and its performance was also analyzed in various environments with different SNR.

Numerical model of compressible gas flow in soil pollution control

Based on the theory of fluid dynamics in porous media, a numerical model of gas flow in unsaturated zone is developed with the consideration of gas density change due to variation of air pressure. This model is characterized of its wider range of availability. The accuracy of this numerical model is analyzed through comparison with modeling results by previous model with presumption of little pressure variation and the validity of this numerical model is shown. Thus it provides basis for the designing and management of landfill gas control system or soil vapor extraction system in soil pollution control.

Steganography Based on BaselineSequential JPEG Compression

Information hiding in Joint Photographic Experts Group （JPEG） compressed images are investigated in this paper. Quantization is the source of information loss in JPEG compression process. Therefore, information hidden in images is probably destroyed by JPEG compression. This paper presents an algorithm to reliably embed information into the JPEG bit streams in the process of JPEG encoding. Information extraction is performed in the process of JPEG decoding. The basic idea of our algorithm is to modify the quantized direct current （DC） coefficients and non zero alternating currenl （AC） coefficients to represent one bit information （0 or 1 ）. Experimental results on gray images using baseline sequential JPEG encoding show that the cover images （images without scoret information） and the stego-images （images with secret information） are perceptually indiscernible.

RESEARCH ON KEY THECHNOLOGIES OF PORNOGRAPHIC IMAGE/VIDEO RECOGNITION IN COMPRESSED DOMAIN

Pornographic image/video recognition plays a vital role in network information surveillance and management. In this paper, its key techniques, such as skin detection, key frame extraction, and classifier design, etc., are studied in compressed domain. A skin detection method based on data-mining in compressed domain is proposed firstly and achieves the higher detection accuracy as well as higher speed. Then, a cascade scheme of pornographic image recognition based on selective decision tree ensemble is proposed in order to improve both the speed and accuracy of recognition. A pornographic video oriented key frame extraction solution in compressed domain and an approach of pornographic video recognition are discussed respectively in the end.

Green Extraction of Bioactive Compounds from Microalgae

At present,there is an increasing demand for natural bioactive compounds able to provide health benefits when included and consumed in a functional food or in a nutraceutical.In this regard,microalgae are promising natural sources with great potential,not only considering that these organisms are largely underexplored,but also because microalgae can be produced at large scale and their chemical composition might be tuned to over-synthesize a particular target compound.The use of advanced sustainable extraction techniques to recover these bioactive compounds is a must nowadays.This work presents an overview on the use of compressed fluid-based extraction techniques to obtain bioactive compounds from microalgae that can be seen also as a first step towards its recovery at larger scale.When relevant,the description of the analytical procedure used to chemically characterize the bioactive compounds is also included.

Image compression based on spatial redundancy removal and image inpainting

We present an algorithm for image compression based on an image inpainting method. First the image regions that can be accurately recovered are located. Then, to reduce the data, information of such regions is removed. The remaining data besides essential details for recovering tile removed regions are encoded to produce output data. At the decoder, an inpainting method is applied to retrieve removed regions using information extracted at the encoder. The image inpainting technique utilizes partial differential equations （PDEs） for recovering information. It is designed to achieve high performance in terms of image compres- sion criteria. This algorithm was examined for various images. A high compression ratio of 1：40 was achieved at an acceptable quality. Experimental results showed attainable visible quality improvement at a high compression ratio compared with JPEG.

青砖茶感官审评方法比较分析

为提出适合青砖茶的感官审评方法,本实验以湖北代表性青砖茶作为研究对象进行煮泡(浸泡式煮茶和喷淋式煮茶),分析不同煮泡条件对青砖茶感官品质及主要品质成分浸出率的影响,并与现行国家标准中冲泡法所得茶汤的感官品质和化学成分进行比较。结果表明,对于紧压型青砖茶,煮泡法茶汤的感官品质明显优于国标冲泡法,煮泡法茶汤的水浸出物、茶多酚、氨基酸、可溶性糖和茶色素等主要品质成分的浸出率远高于国标冲泡法。综合感官品质和品质成分浸出率分析,采用浸泡式煮茶法,在1∶80茶水比、煮沸2 min、静置2~4 min条件下制备的青砖茶茶汤浓度适中,内质品质呈现较为全面。进一步的相关性分析结果表明,可溶性糖、茶褐素及水浸出物浸出率与滋味品质的相关系数较高。该研究结果可为青砖茶及其他类型紧压茶感官品质的客观性评价提供参考。

基于深度卷积测量网络的滚动轴承压缩域故障特征提取方法

压缩感知可有效降低机械状态监测信号的数据存储和传输压力,而现有压缩感知方法在故障诊断的应用中存在压缩效率低下、信号重构过程缓慢等问题。本文利用自编码网络与压缩感知的对应关系,提出了一种基于深度卷积测量网络的滚动轴承压缩域故障特征提取方法。针对无噪声的故障信号样本难以获取的问题,提出一种利用故障机理构建数据集的方法,利用该仿真数据集训练得到的模型适用于不同工况下的实测轴承信号。构造网络层数由所需要的信号压缩率确定、隐含层与原信号的频率呈对应关系的深度卷积去噪自编码网络。截取训练完备的编码子网络(即深度卷积测量网络)代替传统的观测矩阵对滚动轴承振动信号进行压缩测量,实现压缩域的故障特征提取。仿真分析验证了所提数据集构造方法及压缩域特征提取方法的有效性。滚动轴承实验信号分析进一步验证了采用所提方法训练得到的深度卷积测量网络具有很好的泛化性,且能够在压缩率远低于传统压缩感知方法的情况下有效地提取轴承故障特征成分并进行故障诊断。

基于自适应窗口与压缩幅值的瞬时转频估计

针对变转速工况下轴承信号的时频分布能量发散、噪声和谐波干扰强导致其转频难以获取的问题,提出一种基于自适应窗口和压缩幅值的瞬时转频估计方法。首先,通过自适应窗口在频率轴方向搜索出脊线窗口避免噪声和谐波的干扰;其次,在脊线窗口内用压缩幅值方法集中发散的脊线能量;然后,用惩罚函数法提取脊线,实现转频的精确估计;最后,根据采用轴承实验台收集的数据验证了所提出方法的有效性和鲁棒性。结果表明,相比于传统方法,采用所提方法估计瞬时转动频率使误差降低约8%。

基于跨模态引导和对齐的多模态预训练方法

现有的视觉语言多模态预训练方法仅在图像和文本的全局语义上进行特征对齐,对模态间细粒度特征交互的探索不足.针对这一问题,本文提出了一种基于跨模态引导和对齐的多模态预训练方法.该方法在模态特征提取阶段,采用基于视觉序列压缩的双流特征提取网络,在视觉编码器中联合图像和文本信息逐层引导视觉序列压缩,缓解与文本无关的冗余视觉信息对模态间细粒度交互的干扰;在模态特征对齐阶段,对图像和文本特征进行细粒度关系推理,实现视觉标记与文本标记的局部特征对齐,增强对模态间细粒度对齐关系的理解.实验结果表明,本文方法能够更好地对齐视觉文本的细粒度特征,在图文检索任务中,微调后的图像检索和文本检索的平均召回率分别达到了86.4%和94.88%,且零样本图文检索的整体指标相较于经典图文检索算法CLIP(Contrastive Language-Image Pre-training)提升了5.36%,在视觉问答等分类任务中,准确率也优于目前主流多模态预训练方法.

基于云边协同子类蒸馏的卷积神经网络模型压缩方法

当前卷积神经网络模型的训练和分发流程中,云端拥有充足的计算资源和数据集,但难以应对边缘场景中碎片化的需求。边缘侧能够直接进行模型的训练和推理,但难以直接使用云端按照统一规则训练的卷积神经网络模型。针对在边缘侧资源受限的情况下,卷积神经网络算法进行模型压缩的训练和推理有效性低的问题,首先,提出了一种基于云边协同的模型分发和训练框架,该框架可以结合云端和边缘侧各自的优势进行模型再训练,满足边缘对指定识别目标、指定硬件资源和指定精度的需求。其次,基于云边协同框架训练的思路,对知识蒸馏技术进行改进,提出了新的基于Logits和基于Channels两种子类知识蒸馏方法(SLKD和SCKD),云服务端先提供具有多目标识别的模型,而后通过子类知识蒸馏的方法,在边缘侧将模型重新训练为一个可以在资源受限的场景下部署的轻量化模型。最后,在CIFAR-10公共数据集上,对联合训练框架的有效性和两种子类蒸馏算法进行了验证。实验结果表明,在压缩比为50%的情况下,相比具有全部分类的模型,所提模型推理准确率得到了显著的提升(10%~11%);相比模型的重新训练,通过知识蒸馏方法训练出的模型精度也有显著提高,并且压缩比率越高,模型精度提升越明显。

不同温度-围压-气体压力下煤体蠕变-渗流演化规律

为实现深部煤层气的高效开采,通过研究不同温度、围压和气体压力下煤体蠕变变形和渗透率演化规律,得到多因素作用下煤体蠕变-渗流耦合关系;采用自行设计的岩石三轴蠕变-渗流装置,对焦煤进行多因素变量下的压缩蠕变-渗流试验。结果表明:温度与煤样的蠕变呈正相关性,随温度的增加焦煤煤样径向和轴向应变变化速率增大且高温(110℃)下这种变化会一直持续直至煤样破裂;围压强度3 MPa与4 MPa的焦煤煤样在温度30、70、110℃下其气体渗透率降低率最大和最小差值分别为7.8%、5.2%、6.5%和4.2%、2.1%、1.9%;焦煤煤样的渗透率最大降低率随着温度水平升高而增大,试验温度110、70、30℃下的焦煤煤样的气体渗透率最大降低率均值依次为91%、84.6%、73.25%。

银杏叶提取物和经皮球囊压迫术治疗三叉神经痛研究进展

三叉神经痛的病因目前普遍分为两种,即原发性和继发性。原发性三叉神经痛尚无明确具体的发病机制,主要临床症状是面部三叉神经分布区内无固定规律反复发生的剧烈痛感,使患者遭受极大痛苦,生活质量严重降低。目前临床治疗三叉神经痛的主要手术方式中,经皮球囊压迫术能够对患者的疼痛进行有效缓解,是安全、简单、有效的治疗三叉神经痛的首要手术方式。银杏叶提取物富含多种有机酸,能够对三叉神经痛患者的神经功能起到保护和恢复作用。本文综述了经皮球囊压迫术和银杏叶提取物在三叉神经痛治疗中的应用效果,旨在为临床三叉神经痛的治疗提供一定文献依据和参考。

Infant Cry Language Analysis and Recognition:An Experimental Approach

Recently, lots of research has been directed towards natural language processing. However, the baby's cry, which serves as the primary means of communication for infants, has not yet been extensively explored, because it is not a language that can be easily understood. Since cry signals carry information about a babies' wellbeing and can be understood by experienced parents and experts to an extent, recognition and analysis of an infant's cry is not only possible, but also has profound medical and societal applications. In this paper, we obtain and analyze audio features of infant cry signals in time and frequency domains.Based on the related features, we can classify given cry signals to specific cry meanings for cry language recognition. Features extracted from audio feature space include linear predictive coding(LPC), linear predictive cepstral coefficients(LPCC),Bark frequency cepstral coefficients(BFCC), and Mel frequency cepstral coefficients(MFCC). Compressed sensing technique was used for classification and practical data were used to design and verify the proposed approaches. Experiments show that the proposed infant cry recognition approaches offer accurate and promising results.

基于声纹压缩的变压器状态检测技术研究

为提高变压器运维的智能化水平,提出一种基于声纹压缩的变压器状态检测新方法。该方法利用声纹特征提取技术从变压器运行声音中提取关键特征,应用声纹压缩算法实现高效编码,并基于压缩后的特征训练状态识别模型。实验结果表明,该方法在故障检测率、虚警率、准确率等指标上均优于传统方法,可有效提升变压器状态检测的效率和可靠性。

带压注浆封孔用聚氨酯材料性能研究

为解决突出矿井煤层抽采钻孔瓦斯抽采衰减速度快,封孔漏气等问题,采取带压封孔工艺技术,经过试验对比,选定密度为164.5 kg/m^(3)的聚氨酯A、B料作为带压封孔材料。经对比实验验证:带压封孔工艺将瓦斯抽采浓度提升到34.2%~57.9%,单孔抽采量达到796.15~982.63 m^(3),瓦斯抽采效果明显提高。