摘要：【中文】本发明实施例提供了一种图像处理方法及装置，该图像处理方法包括：将N帧图像输入到质量评估网络模型，利用质量评估网络模型，根据M个质量评估参数确定N帧图像的最终质量评估值，根据N帧图像的最终质量评估值，确定目标图像或目标向量。通过本发明实施例提供的技术方案，基于M个质量评估参数对N帧图像进行评估，得到N帧图像的最终质量评估值，再根据最终质量评估值确定目标图像。对于N帧图像中的每一帧图像均是根据M个质量评估参数进行评估的，通过该M个质量评估参数提高了每一帧图像的图像质量评估准确率。最终质量评估值根据M个质量评估参数所得到的，对于根据最终质量评估值所确定的目标图像，提高了最终输出的目标图像的成像质量。 【EN】The embodiment of the invention provides an image processing method and device, wherein the image processing method comprises the following steps: inputting the N frames of images into a quality evaluation network model, determining a final quality evaluation value of the N frames of images according to the M quality evaluation parameters by using the quality evaluation network model, and determining a target image or a target vector according to the final quality evaluation value of the N frames of images. According to the technical scheme provided by the embodiment of the invention, the N frames of images are evaluated based on M quality evaluation parameters to obtain the final quality evaluation value of the N frames of images, and then the target image is determined according to the final quality evaluation value. And evaluating each frame of image in the N frames of images according to the M quality evaluation parameters, and improving the image quality evaluation accuracy of each frame of image through the M quality evaluation parameters. The final quality evaluation value is obtained according to the M quality evaluation parameters, and the imaging quality of the finally output target image is improved for the target image determined according to the final quality evaluation value.

摘要：【中文】本发明实施例提供了一种用于目标识别的图像处理方法及装置，方法包括：将N帧图像输入到质量评估网络模型，利用质量评估网络模型，根据预设的子区域图像的关注权重和M个质量评估参数确定N帧图像中每一帧图像的特征向量，根据N帧图像的特征向量确定N帧图像的最终质量评估值；根据N帧图像的最终质量评估值确定用于进行目标识别的目标图像或目标向量。通过本发明实施例提供的技术方案，基于子区域图像的关注权重和M个质量评估参数，确定每帧图像的特征向量。每帧图像均根据子区域图像的关注权重和M个质量评估参数进行评估的，提高了每帧图像的图像质量评估准确率。最终质量评估值是根据图像的特征向量得到的，提高了目标图像的成像质量。 【EN】The embodiment of the invention provides an image processing method and device for target identification, wherein the method comprises the following steps: inputting the N frames of images into a quality evaluation network model, determining a characteristic vector of each frame of image in the N frames of images according to the attention weight of a preset subregion image and M quality evaluation parameters by using the quality evaluation network model, and determining a final quality evaluation value of the N frames of images according to the characteristic vector of the N frames of images; and determining a target image or a target vector for target identification according to the final quality evaluation value of the N frames of images. According to the technical scheme provided by the embodiment of the invention, the characteristic vector of each frame of image is determined based on the attention weight of the sub-region image and M quality evaluation parameters. Each frame of image is evaluated according to the attention weight of the sub-region image and the M quality evaluation parameters, so that the image quality evaluation accuracy of each frame of image is improved. The final quality evaluation value is obtained according to the feature vector of the image, and the imaging quality of the target image is improved.

摘要：【中文】本申请公开了一种基于邻域像素关系的边缘像素点提取方法及装置，所述方法包括：获取待检测图像；在所述待检测图像中确定待检测区域；获取所述待检测区域中第i像素点的邻域像素点之灰度值；根据所述第i像素点的邻域像素点之灰度值确定边缘像素点，所述方法及装置仅基于邻域像素点的灰度关系来确认该像素点是否为边缘像素点，避免设定阈值，从而解决由于阈值设定不适当而造成的边缘像素点漏检或者误检的问题。 【EN】The application discloses a method and a device for extracting edge pixel points based on neighborhood pixel relationship, wherein the method comprises the following steps: acquiring an image to be detected; determining a region to be detected in the image to be detected; acquiring the gray value of a neighborhood pixel point of the ith pixel point in the area to be detected; the method and the device determine the edge pixel point according to the gray value of the neighborhood pixel point of the ith pixel point, and confirm whether the pixel point is the edge pixel point only based on the gray value relation of the neighborhood pixel point, so that the threshold value is prevented from being set, and the problem of missing detection or false detection of the edge pixel point caused by improper setting of the threshold value is solved.

摘要：【中文】本发明公开的一种多用途画箱，包括储物箱，储物箱顶端一侧活动连接有画板，储物箱的两侧面旋转连接有折叠搁物台，储物箱的底部连接有伸缩支撑架。通过画板与储物箱活动连接，能够打开储物箱用于防止绘画工具；调节支架的使用能够调节画板的高度与角度；折叠搁物台绘画时用以放置绘画工具；伸缩支撑架既能作为桌腿使用，又能用以移动绘画箱。本发明的一种多用途画箱，具有多种用途，既能容纳多种绘画工具，作为绘画支架使用，又能方便携带，使用用途多样。 【EN】The invention discloses a multipurpose drawing box which comprises a storage box, wherein one side of the top end of the storage box is movably connected with a drawing board, two side surfaces of the storage box are rotatably connected with a folding object placing table, and the bottom of the storage box is connected with a telescopic support frame. The drawing board is movably connected with the storage box, so that the storage box can be opened to prevent drawing tools; the height and the angle of the drawing board can be adjusted by using the adjusting bracket; the folding shelf is used for placing a drawing tool when drawing; the telescopic support frame can be used as a table leg and can be used for moving the drawing box. The multipurpose drawing box has multiple purposes, can accommodate various drawing tools to be used as a drawing support, is convenient to carry, and has multiple use purposes.

摘要：【中文】本发明涉及一种读出电路包括一个或多个像素读出模块，配置为将光电二极管输出的电流信号转化为电压信号，每个所述像素读出模块包括第一源跟随器；一个或多个列级传输模块，其中每个所述列级传输模块包括第二源跟随器；输出级传输模块，配置为对所述一个或多个列级传输模块的输出信号进行缓冲；其中所述输出级传输模块电连接在其输出端与第一输入端之间的、与所述第一源跟随器相应的第一校准单元和与所述第二源跟随器相应的第二校准单元，二者共同配置为对所述输出级传输模块的输出信号进行校准以消除所述第一和第二源跟随器引入的偏差。本发明还涉及一种探测设备。 【EN】The invention relates to a readout circuit, which comprises one or more pixel readout modules, a first source follower and a second source follower, wherein the pixel readout modules are configured to convert current signals output by photodiodes into voltage signals; one or more column level transmission modules, wherein each of the column level transmission modules comprises a second source follower; an output stage transmission module configured to buffer output signals of the one or more column stage transmission modules; and the output stage transmission module is electrically connected between the output end and the first input end of the output stage transmission module, and comprises a first calibration unit corresponding to the first source follower and a second calibration unit corresponding to the second source follower, and the first calibration unit and the second calibration unit are jointly configured to calibrate the output signal of the output stage transmission module so as to eliminate the deviation introduced by the first source follower and the second source follower. The invention also relates to a detection device.

摘要：【中文】本申请公开了一种信号处理装置以及方法。信号处理装置包括：检测单元，配置为基于注入电压信号生成检测信号，其中，检测信号用于指示注入电压信号的幅值区间；驱动单元，配置为基于检测信号生成驱动信号；输出单元，配置为接收注入电压信号，并基于驱动信号来在与幅值区间相对应的输出端处输出注入电压信号。 【EN】The application discloses a signal processing device and a method. The signal processing apparatus includes: a detection unit configured to generate a detection signal based on the injection voltage signal, wherein the detection signal is used for indicating an amplitude interval of the injection voltage signal; a driving unit configured to generate a driving signal based on the detection signal; an output unit configured to receive the injection voltage signal and output the injection voltage signal at an output terminal corresponding to the amplitude interval based on the driving signal.

摘要：【中文】本发明公开了一种基于深度学习的单能CT能谱成像方法，包括以下步骤：以单能CT图像和标签高能CT图像作为一个训练样本，组成训练集；基于深度学习网络构建用于根据输入的单能CT图像预测输出高能CT图像的图像转换模型，并利用训练集训练该图像转换模型以确定图像转换模型的模型参数；将需要成像的单能CT图像输入至模型参数确定的图像转换模型中，进计算输出高能CT图像。该方法可以实现使用单能CT图像快速、准确和鲁棒地获得高能CT图像，满足临床的需求。 【EN】The invention discloses a single-energy CT energy spectrum imaging method based on deep learning, which comprises the following steps of: taking the single-energy CT image and the label high-energy CT image as a training sample to form a training set; constructing an image conversion model for predicting and outputting a high-energy CT image according to an input single-energy CT image based on a deep learning network, and training the image conversion model by utilizing a training set to determine model parameters of the image conversion model; and inputting the single-energy CT image to be imaged into an image conversion model determined by the model parameters, and calculating to output a high-energy CT image. The method can realize the rapid, accurate and robust acquisition of the high-energy CT image by using the single-energy CT image, and meets the clinical requirements.

摘要：【中文】本发明提供一种视频分类方法、装置及设备、存储介质，视频分类方法，包括：将待分类处理的视频帧序列输入至已训练的神经网络中，以由所述神经网络对所述视频帧序列中每N个视频帧在目标域上进行特征关联得到对应的第一关联特征向量，对所有第一关联特征向量在时间域上进行特征关联得到第二关联特征向量，并对所述第二关联特征向量进行分类得到并输出分类结果；依据所述神经网络输出的分类结果确定所述视频帧序列的类别；其中，所述N为小于所述视频帧序列长度的正整数，所述目标域至少包括空间域。改善对视频在时间域和空间域上的信息表达，有利于提高视频分类准确度。 【EN】The invention provides a video classification method, a video classification device, video classification equipment and a storage medium, wherein the video classification method comprises the following steps: inputting a video frame sequence to be classified into a trained neural network, performing feature association on every N video frames in the video frame sequence by the neural network on a target domain to obtain corresponding first associated feature vectors, performing feature association on all the first associated feature vectors on a time domain to obtain second associated feature vectors, classifying the second associated feature vectors to obtain a classification result, and outputting the classification result; determining the category of the video frame sequence according to the classification result output by the neural network; wherein N is a positive integer less than the length of the video frame sequence, and the target domain includes at least a spatial domain. The information expression of the video in the time domain and the space domain is improved, and the video classification accuracy is improved.

摘要：【中文】本发明公开了一种基于类凸集投影算法的图像恢复算法，用于解决现有图像恢复网络特征图中存在大量噪声的问题。实现步骤为：1)构建训练样本集和测试样本集；2)构建显示的特征去噪模块；3)构建基于类凸集投影算法的图像恢复网络模型PL‑DSR；4)对PL‑DSR网络模型进行训练；5)对PL‑DSR网络模型进行测试；6)生成恢复图像。本发明基于传统的凸集投影算法构建一种显示的特征去噪模块，用于对特征图中的噪声进行抑制，将其引入到经典EDSR网络中，构建出基于类凸集投影算法的图像恢复网络PL‑DSR。该网络可以有效的抑制特征图中的噪声，提取到结构化更加清晰的特征，在减少模型复杂度的同时进一步提升了恢复效果和人的视觉感官体验，可用于完成摄影、影视作品的后期制作加工。 【EN】The invention discloses an image recovery algorithm based on a convex set projection algorithm, which is used for solving the problem that a large amount of noise exists in a characteristic graph of the conventional image recovery network. The method comprises the following implementation steps: 1) constructing a training sample set and a testing sample set; 2) constructing a displayed characteristic denoising module; 3) constructing an image recovery network model PL-DSR based on a convex set projection algorithm; 4) training a PL-DSR network model; 5) testing a PL-DSR network model; 6) a restored image is generated. The invention constructs a displayed feature denoising module based on the traditional convex set projection algorithm, is used for inhibiting noise in a feature map, introduces the noise into a classic EDSR network, and constructs an image recovery network PL-DSR based on the convex set projection algorithm. The network can effectively inhibit noise in the characteristic diagram, extracts the characteristics with clearer structurization, further improves the recovery effect and the visual sense experience of people while reducing the complexity of the model, and can be used for finishing post-production and processing of photography and film and television works.

摘要：【中文】本发明涉及图像处理技术领域，具体是一种展示图片生成方法、装置及存储介质，所述方法包括：获取原始图片和图片目标尺寸；根据所述原始图片和所述图片目标尺寸确定所述原始图片所对应的展示图片的待填充区域；读取所述原始图片中的像素点信息；根据所述像素点信息确定所述原始图片中像素点颜色的离散度；依据所述离散度选取所述原始图片所对应的展示图片的图片填充方式；按照所述图片填充方式对所述待填充区域进行填充，生成所述展示图片。本发明的展示图片生成方法通过根据原始图片中像素点颜色的离散度来选取展示图片的图片填充方式，并对所述展示图片的待填充区域进行填充，使得填充后的展示图片颜色过渡更加自然，提高图片的展示效果。 【EN】The invention relates to the technical field of image processing, in particular to a display picture generation method, a display picture generation device and a storage medium, wherein the method comprises the following steps: acquiring an original picture and a picture target size; determining a region to be filled of a display picture corresponding to the original picture according to the original picture and the picture target size; reading pixel point information in the original picture; determining the dispersion of pixel point colors in the original picture according to the pixel point information; selecting a picture filling mode of a display picture corresponding to the original picture according to the dispersion; and filling the area to be filled according to the picture filling mode to generate the display picture. According to the display picture generation method, the picture filling mode of the display picture is selected according to the dispersion of the pixel point colors in the original picture, and the region to be filled of the display picture is filled, so that the color transition of the filled display picture is more natural, and the display effect of the picture is improved.