摘要：【中文】本发明公开了一种清热消肿的中草药膏。本发明由以下重量份的组分组成：牡丹皮20‑35份、黄芪30‑40份、白及25‑45份、防己15‑35份、白术45‑55份、三七20‑40份、五倍子25‑45份、地榆25‑45份、当归30‑45份、黄凡士林25‑55份、羊毛脂15‑35份、羟苯甲基9‑28份、水45‑75份；本发明清热消肿的中草药膏具有清热燥湿、活血消肿的功效，对痔疮的肿痛可以有效的消肿生肌、止血、止痛，适用于痔疾肿痛，肛裂的人群。 【EN】The invention discloses a Chinese herbal medicine ointment for clearing heat and diminishing swelling. The invention comprises the following components in parts by weight: 20-35 parts of tree peony bark, 30-40 parts of astragalus, 25-45 parts of bletilla, 15-35 parts of radix stephaniae tetrandrae, 45-55 parts of bighead atractylodes rhizome, 20-40 parts of pseudo-ginseng, 25-45 parts of Chinese gall, 25-45 parts of garden burnet, 30-45 parts of angelica, 25-55 parts of yellow vaseline, 15-35 parts of lanolin, 9-28 parts of hydroxybenzyl and 45-75 parts of water; the Chinese herbal medicine ointment for clearing heat and diminishing swelling has the effects of clearing heat and drying dampness, and activating blood and diminishing swelling, can effectively reduce swelling and promote tissue regeneration, stop bleeding and relieve pain for swelling and pain of hemorrhoids, and is suitable for people suffering from hemorrhoid swelling and pain and anal fissure.

摘要：【中文】本发明公开了一种清热消肿的中草药组合物。本发明由牡丹皮、黄芪、白及、防己、白术、三七、五倍子、地榆和当归组成。本发明所选用的组分具有清热燥湿、活血消肿的功效，对痔疮的肿痛可以有效的消肿生肌、止血、止痛，适用于痔疾肿痛，肛裂的人群。 【EN】The invention discloses a Chinese herbal medicine composition for clearing heat and diminishing swelling. The invention is composed of tree peony bark, astragalus root, bletilla, tetrandra root, white atractylodes rhizome, notoginseng, gallnut, sanguisorba root and angelica. The components selected by the invention have the effects of clearing heat and drying dampness, and activating blood and reducing swelling, can effectively reduce swelling and promote granulation, stop bleeding and relieve pain for swelling and pain of hemorrhoids, and is suitable for people suffering from hemorrhoid swelling and pain and anal fissure.

摘要：【中文】本发明公开了一种基于人工智能的预警系统，包括远程监控中心、监控数据采集终端、火灾预测模块、火情判别模块和无线通信模块，监控终端、火灾预测模块和火情判别模块通过无线通信模块与远程监控中心连接，远程控制中心包括显示模块、中心计算机、历史火情知识库、应急方案数据库和在线报警模块；本发明系统能够通过监控数据采集终端采集监控区域内的各种数据，通过在CCD相机前加上滤光片能够有效减少背景干扰，进而提高火灾识别效率，通过火灾预测模块和火情判别模块，根据可能发生的火灾属于人工可扑救成功率的影响级别进行后续处理，能够为远程监控中心提供更为精准的实时数据，能便于远程监控中心提供更为精准科学、安全的预警决策。 【EN】The invention discloses an artificial intelligence-based early warning system which comprises a remote monitoring center, a monitoring data acquisition terminal, a fire prediction module, a fire judgment module and a wireless communication module, wherein the monitoring terminal, the fire prediction module and the fire judgment module are connected with the remote monitoring center through the wireless communication module; the system can acquire various data in a monitoring area through the monitoring data acquisition terminal, can effectively reduce background interference by adding the optical filter in front of the CCD camera, further improves the fire identification efficiency, performs subsequent processing according to the influence level of the success rate of the fire which can be suppressed manually through the fire prediction module and the fire distinguishing module, can provide more accurate real-time data for a remote monitoring center, and can provide more accurate, scientific and safe early warning decision for the remote monitoring center.

摘要：【中文】本发明公开了一种基于深度学习的推荐方法，包括以下步骤：对推荐项目的内容进行划分，将文本信息进行文本结构统一；对内容文本进行数据预处理，具体包括对内容文本进行词组拆分、语义提取、语义延伸和词组重组；利用训练好的深度神经网络模型对经过数据预处理的内容文本进行特征提取；对提取的特征进过滤处理，然后基于过滤处理后的特征进行内容推荐；本发明通过对推荐内容的文本信息进行多级处理，能够对文本信息的语义内容进行精确深度获取，然后利用深度神经网络模型对深度化的内容文本进行训练，可以得出最优输出，保证推荐内容丰富度和准确性高。 【EN】The invention discloses a deep learning-based recommendation method, which comprises the following steps: dividing the content of the recommended item, and unifying the text structure of the text information; performing data preprocessing on the content text, specifically including performing phrase splitting, semantic extraction, semantic extension and phrase recombination on the content text; performing feature extraction on the content text subjected to data preprocessing by using the trained deep neural network model; filtering the extracted features, and then recommending contents based on the filtered features; according to the method, the text information of the recommended content is subjected to multi-stage processing, the semantic content of the text information can be accurately and deeply acquired, then the deep neural network model is used for training the deep content text, the optimal output can be obtained, and the richness and the high accuracy of the recommended content are guaranteed.

摘要：【中文】本发明涉及可控滑差机械离合器，包括行星架一，其内侧安装有一对行星齿轮，与两个行星齿轮同时啮合安装有半轴齿轮，两个半轴齿轮中部均安装有半轴，行星架一外侧面通过外齿轮一啮合有齿轮一；齿轮一及两个半轴中一个为动力输入进行传递，且其中任意一个、两个或三个连接有液力控制系统。或者，包括行星架二，其上的多个行星轮外侧共同啮合有齿圈，多个行星轮内侧共同啮合有太阳轮，齿圈外侧面通过外齿轮二啮合有齿轮三；齿轮三、传动轴一和传动轴二中一个为动力输入进行传递，且其中任意一个、两个或三个连接有液力控制系统。本发明具备现有离合器的节能优点，并兼备无级调节滑差的功能，以保证车辆的平稳起步和换挡平顺。 【EN】The invention relates to a controllable slip mechanical clutch, which comprises a first planet carrier, wherein a pair of planet gears are arranged on the inner side of the first planet carrier, half shaft gears are simultaneously meshed with the two planet gears, half shafts are arranged in the middle parts of the two half shaft gears, and a first gear is meshed with the outer side surface of the first planet carrier through a first outer gear; one of the first gear and the two half shafts is used for power input transmission, and any one, two or three of the first gear and the two half shafts are connected with a hydraulic control system. Or the outer side of the gear ring is meshed with a gear III through an outer gear II; one of the gear III, the transmission shaft I and the transmission shaft II is used for power input transmission, and any one, two or three of the gears are connected with a hydraulic control system. The invention has the energy-saving advantage of the existing clutch and has the function of stepless slip regulation so as to ensure the stable starting and smooth gear shifting of the vehicle.

摘要：【中文】本发明属于图像增强技术领域，公开了一种基于非线性导向滤波的图像增强方法，基于非线性导向滤波的图像增强系统包括：图像采集模块、主控模块、预处理模块、分割模块、滤波构建模块、滤波系数计算模块、降噪模块、图像增强模块、图像输出模块、显示模块。本发明提供了一种一种基于非线性导向滤波的图像增强方法，可以进一步提高效率，减少时间复杂度，应用很多不同的计算机视觉领域，较好保持边缘的图象平滑，图像细节增强，提高图像的质量，增强图像的显示效果，HDR图像压缩及去除均匀雾等操作，该算法计算简便，高效；通过分割模块能够极大提升图像分割的精度和速度；同时，通过降噪模块可以高效降噪，大大提高图像质量。 【EN】The invention belongs to the technical field of image enhancement, and discloses an image enhancement method based on nonlinear guide filtering, wherein an image enhancement system based on nonlinear guide filtering comprises the following steps: the device comprises an image acquisition module, a main control module, a preprocessing module, a segmentation module, a filtering construction module, a filtering coefficient calculation module, a noise reduction module, an image enhancement module, an image output module and a display module. The invention provides an image enhancement method based on nonlinear guided filtering, which can further improve the efficiency, reduce the time complexity, be applied to a plurality of different computer vision fields, better keep the image smoothness of the edge, enhance the image details, improve the image quality, enhance the image display effect, compress HDR images, remove uniform fog and other operations, and has simple and efficient algorithm calculation; the image segmentation precision and speed can be greatly improved through the segmentation module; meanwhile, the noise reduction module can be used for efficiently reducing noise, and the image quality is greatly improved.

摘要：【中文】本发明涉及叶包茶制作技术领域，且公开了叶包茶制作工艺以及加工工装，包括以下步骤：步骤一：把植物叶片或者树叶放入清洗设备中清洗干净；步骤二：把符合标准加工好的茶叶或养生茶原料包裹到叶子中间与绿叶为一体；步骤三：把用叶子包好的茶叶或养生茶原料放到烘干机烘干，进行备用；步骤四：把包裹的茶叶或养生茶原料放入杯中，再开几个孔，用100度的开水冲泡，从而制得叶包茶。该叶包茶制作工艺以及加工工装，在生产茶叶或者花茶的过程中，加上植物或者树叶包茶这个制做工艺方法，把茶包在叶子当中，能使茶味道保持更香，提高绿叶中叶绿素的食用和药用价值，叶子包起来的茶也更好看美观。 【EN】The invention relates to the technical field of tea bag making, and discloses a tea bag making process and a processing tool, which comprise the following steps: the method comprises the following steps: putting plant leaves or leaves into cleaning equipment to be cleaned; step two: wrapping the processed tea or health tea raw materials meeting the standard into the middle of the leaves to integrate with the green leaves; step three: drying the tea leaves or the health tea raw materials wrapped by the leaves in a dryer for later use; step four: the wrapped tea leaves or the health tea raw materials are put into a cup, a plurality of holes are formed, and the tea is brewed by boiled water with the temperature of 100 ℃, so that the tea bag is prepared. According to the leaf-wrapped tea making process and the processing tool, in the process of producing tea or scented tea, the tea is wrapped in the leaves by adding the plant or leaf-wrapped tea making process method, so that the taste of the tea can be kept more fragrant, the edible and medicinal values of chlorophyll in green leaves are improved, and the tea wrapped in the leaves is better in appearance.

摘要：【中文】本发明公开的属于数码设备图像处理技术领域，具体为一种便携式数码设备的图像处理方法，该图像处理方法如下：步骤一：图像收集单元收集光信号信息，即将图像的输入端对外部环境光信号进行收集；步骤二：图像信息通过光电转换单元进行光电转换；步骤三：转换成模拟信号后输出到图像处理单元中；步骤四：图像处理单元对图像电信号进行处理；步骤五：在图像处理单元对图像处理的同时，中央控制器对图像电信号中的数据进行优化，通过图像处理单元对图像信息进行处理优化的同时，通过中央控制器进行辅助处理，减少图像处理单元的工作负载，从而提高了图像处理单元的图像处理速度，减少高负载下的高温情况出现，提高装置的使用寿命。 【EN】The invention discloses an image processing method of portable digital equipment, belonging to the technical field of image processing of digital equipment, comprising the following steps: the method comprises the following steps: the image collecting unit collects light signal information, namely collecting an external environment light signal by an input end of an image; step two: the image information is subjected to photoelectric conversion through a photoelectric conversion unit; step three: converting the analog signal into an analog signal and outputting the analog signal to an image processing unit; step four: the image processing unit processes the image electric signal; step five: when the image processing unit processes the image, the central controller optimizes data in the image electric signal, and when the image processing unit processes and optimizes the image information, the central controller performs auxiliary processing to reduce the workload of the image processing unit, thereby improving the image processing speed of the image processing unit, reducing the occurrence of high temperature under high load and prolonging the service life of the device.

摘要：【中文】本发明公开了一种基于去磁电感滤波磁路的高压直流输电线感应取能装置及方法。通过在CT铁芯的第二磁路分支上绕制去磁线圈及并联去磁电感，以及在第三磁路分支上绕制取能线圈及并联负载和等效助磁电容，实现了脉动磁通分量在第三磁路分支中的分流。进一步通过在第三磁路分支上增加抑制直流磁通单元（开气隙），实现了直流磁通分量在第二磁路分支中的分流。根据感应取能装置的总体结构，建立等效磁路模型，进而建立第三磁路分支中脉动磁通以及直流磁通的数学模型，通过改进感应取能CT的铁芯构造，增加了去磁线圈、去磁电感和等效助磁电容，在获取相同能量的情况下，获取了所需的能量，同时减小工程实现的代价。 【EN】The invention discloses an induction energy-taking device and method for a high-voltage direct-current transmission line based on a demagnetizing inductor filtering magnetic circuit. The shunt of the pulsating magnetic flux component in the third magnetic circuit branch is realized by winding the demagnetization coil and the parallel demagnetization inductor on the second magnetic circuit branch of the CT iron core, and winding the energy taking coil, the parallel load and the equivalent magnetic assisting capacitor on the third magnetic circuit branch. By further adding a dc flux suppressing unit (open air gap) to the third magnetic circuit branch, a shunting of the dc flux component in the second magnetic circuit branch is achieved. According to the overall structure of the induction energy taking device, an equivalent magnetic circuit model is established, a mathematical model of pulsating magnetic flux and direct-current magnetic flux in a third magnetic circuit branch is further established, a demagnetizing coil, a demagnetizing inductor and an equivalent magnetic assisting capacitor are added by improving the iron core structure of the induction energy taking CT, required energy is obtained under the condition of obtaining the same energy, and meanwhile, the cost of engineering realization is reduced.

摘要：【中文】本发明公开了一种图像清晰度处理方法，包括以下步骤：根据原图像主体类型确定强调幅度和门限值；根据原图像幅面尺寸和分辨率参数选择半径值；对原图像包括的不同色版通道进行逐一的清晰度强调处理；清晰度处理，得到清晰图像；对清晰图像进行灰度调整，然后对清晰图像进行分层处理，分为高频图层和低频图层，分别对高频图层和低频图层基于小波变换的降噪处理，得到新的清晰图像；本发明方法通过根据图像的不同特点和类型，合理的选择强调幅度、半径值和门限值，能够在各功能综合平衡使用的基础上适当地进行最适宜的清晰度调节，进行基于小波变换的降噪处理可以针对性的降噪，新的清晰图像能够进一步稳定的保持图像信息和清晰度的完整性。 【EN】The invention discloses an image definition processing method, which comprises the following steps: determining an emphasis amplitude and a threshold value according to the type of the original image main body; selecting a radius value according to the breadth size and the resolution parameter of an original image; performing definition emphasis processing on different color plate channels included in an original image one by one; performing definition processing to obtain a clear image; carrying out gray level adjustment on the clear image, then carrying out layering processing on the clear image, dividing the clear image into a high-frequency layer and a low-frequency layer, and respectively carrying out noise reduction processing on the high-frequency layer and the low-frequency layer based on wavelet transform to obtain a new clear image; the method of the invention can properly adjust the most suitable definition on the basis of the comprehensive balanced use of all functions by reasonably selecting the emphasis amplitude, the radius value and the threshold value according to different characteristics and types of the image, can perform targeted noise reduction by performing noise reduction processing based on wavelet transformation, and can further stably maintain the integrity of image information and definition by a new clear image.