摘要：【中文】本申请提供一种信息筛选方法、装置、电子设备及可读存储介质，包括：获取问题以及与问题对应的一个答案选项组成的信息组合，问题对应有多个答案选项，一个答案选项为多个答案选项中的任一个，多个答案选项中的至少一个答案选项为问题的正确答案；使用训练完成的信息筛选模型对信息组合进行分析处理，得到信息组合的判定结果。本申请实施例提供的信息筛选方法可以利用信息筛选模型来判断选择题中的题干以及其中一个选项组成的组合的正确或错误，从而获得上述选择题的正确答案。由于本申请实施例是利用训练完成的信息筛选模型来解答选择题，因此在对题库依赖较小的情况下依然能得到选择题的正确答案，能够更好地对学生群体进行习题辅导。 【EN】The application provides an information screening method, an information screening device, an electronic device and a readable storage medium, wherein the information screening method comprises the following steps: acquiring an information combination consisting of a question and one answer option corresponding to the question, wherein the question corresponds to a plurality of answer options, one answer option is any one of the answer options, and at least one answer option in the answer options is a correct answer to the question; and analyzing and processing the information combination by using the trained information screening model to obtain a judgment result of the information combination. The information screening method provided by the embodiment of the application can judge the correctness or the mistake of the question stem in the choice questions and the combination of one option by utilizing the information screening model, thereby obtaining the correct answer of the choice questions. Because the embodiment of the application utilizes the trained information screening model to solve the choice questions, the correct answers of the choice questions can still be obtained under the condition of less dependence on the question library, and the student group can be better guided by exercises.

摘要：【中文】本发明提供了一种针对CT影像数据的目标检测方法、装置及电子设备，涉及图像处理技术领域，获取CT影像数据；CT影像数据包括目标特征；将CT影像数据进行像素归一化，得到多个灰度图像；在多个灰度图像中按照预设的方法提取灰度图像组，并将每个灰度图像组合并为对应的伪彩色图像；灰度图像组为灰度图像按照指定数量和指定顺序组成的；将伪彩色图像输入预先训练好的神经网络中，以使预先训练好的神经网络对伪彩色图像进行检测，得到目标特征的类别及位置。因此本发明可以有效提高CT影像数据目标检测的准确率。 【EN】The invention provides a target detection method, a target detection device and electronic equipment for CT image data, relates to the technical field of image processing, and aims to obtain the CT image data; the CT image data includes target features; carrying out pixel normalization on the CT image data to obtain a plurality of gray level images; extracting a gray image group from a plurality of gray images according to a preset method, and combining each gray image group into a corresponding pseudo color image; the gray image group is composed of gray images according to the specified number and the specified sequence; and inputting the pseudo-color image into a pre-trained neural network so that the pre-trained neural network detects the pseudo-color image to obtain the category and the position of the target feature. Therefore, the method can effectively improve the accuracy of the target detection of the CT image data.

摘要：【中文】本发明提供了一种病灶检测方法、装置及电子设备，该方法获取待检测的医学扫描图像；对医学扫描图像进行预处理，得到与医学扫描图像对应的去噪扫描图像；将去噪扫描图像输入至预先训练的病灶检测模型，得到病灶检测模型输出的检测结果；其中，检测结果包括病灶中心点坐标、病灶尺寸和病灶偏移量中的一种或多种；将检测结果标注至医学扫描图像。本发明可以有效提高检测病灶的准确度。 【EN】The invention provides a focus detection method, a focus detection device and electronic equipment, wherein the method is used for acquiring a medical scanning image to be detected; preprocessing the medical scanning image to obtain a de-noised scanning image corresponding to the medical scanning image; inputting the de-noised scanning image into a pre-trained focus detection model to obtain a detection result output by the focus detection model; wherein, the detection result comprises one or more of the coordinates of the focus central point, the focus size and the focus offset; and labeling the detection result to the medical scanning image. The invention can effectively improve the accuracy of detecting the focus.

摘要：【中文】本发明提供了一种病灶检测系统构建的方法、装置及电子设备，包括：获取CT影像数据；其中，CT影像数据包括有病灶的CT影像数据和无病灶的CT影像数据；基于类空间金字塔检测网络模型对CT影像数据进行逐层检测，确定病灶类型和病灶位置；根据病灶类型和病灶位置构建病灶检测模型；基于病灶检测模型构建病灶检测系统；其中，病灶检测系统包括至少两个病灶检测模型。本发明可以基于类空间金字塔检测网络模型构建病灶检测系统，在构建过程中无需对数据进行肺实质分割等预处理，也无需通过后处理降低误差率，从而可以简化系统的构建过程。 【EN】The invention provides a method and a device for constructing a focus detection system and electronic equipment, wherein the method comprises the following steps: acquiring CT image data; wherein, the CT image data comprises CT image data with focus and CT image data without focus; detecting the CT image data layer by layer based on a space pyramid detection network model to determine the type and position of a focus; constructing a focus detection model according to the focus type and the focus position; constructing a focus detection system based on a focus detection model; wherein, the focus detection system comprises at least two focus detection models. The invention can construct a focus detection system based on the space pyramid-like detection network model, and the lung parenchyma segmentation and other preprocessing of data are not needed in the construction process, and the error rate is also not needed to be reduced through post-processing, so that the construction process of the system can be simplified.

摘要：【中文】本发明提供了一种基于度量学习的菜品识别方法及装置，涉及图像识别和处理的技术领域，包括：先获取待检测图像；若待检测图像为菜品图像，则对待检测图像进行裁剪，得到包含菜品信息的目标图像；然后将目标图像输入至目标卷积神经网络中，得到目标图像的特征信息；目标卷积神经网络为基于度量学习训练的网络；再利用最近邻方法从预设数据库中选择与目标图像的特征信息相似的图像特征信息；最后将图像特征信息对应的菜品信息作为待检测图像的菜品信息。本发明基于度量学习训练好的卷积神经网络模型对菜品图像的识别粒度小，进而可以区分特征相似的菜品。度量学习还可以通过优化特征空间，实现对类内距离大的菜品图像的识别。 【EN】The invention provides a dish identification method and device based on metric learning, which relate to the technical field of image identification and processing and comprise the following steps: firstly, acquiring an image to be detected; if the image to be detected is a dish image, cutting the image to be detected to obtain a target image containing dish information; then inputting the target image into a target convolutional neural network to obtain the characteristic information of the target image; the target convolutional neural network is a network based on metric learning training; selecting image characteristic information similar to the characteristic information of the target image from a preset database by using a nearest neighbor method; and finally, using the dish information corresponding to the image characteristic information as the dish information of the image to be detected. The method is based on the convolution neural network model which is well trained by metric learning and has small identification granularity on dish images, so that dishes with similar characteristics can be distinguished. The measurement learning can also realize the identification of the dish images with large intra-class distance by optimizing the feature space.