摘要：【中文】本发明公开了一种可移动式升降高光墙，属于汽车造型模型高光检验设备技术领域，针对目前的检测灯幕所存在的诸多问题，可移动式升降高光墙，由支架、灯幕、导向机构、传动机构、配电箱和手电门组成；灯幕在导向机构的导向作用和传动机构的牵引驱动下能够实现灯幕的升降，能够调整等距平行光源在不同空间的位置和各种姿态，该升降高光墙底部设有脚轮，能够方便移动，实现对汽车造型模型各个型面的高光检测。 【EN】The invention discloses a movable lifting highlight wall, which belongs to the technical field of highlight inspection equipment of automobile modeling models and aims at various problems of the existing detection lamp curtain, and the movable lifting highlight wall consists of a bracket, a lamp curtain, a guide mechanism, a transmission mechanism, a distribution box and a flashlight door; the lamp curtain can realize the lift of lamp curtain under guiding mechanism's guiding action and drive mechanism's traction drive, can adjust the position and various gestures of equidistance parallel light source in different spaces, and this lift highlight wall bottom is equipped with the truckle, can conveniently remove, realizes the highlight detection to each profile of automobile modeling model.

摘要：【中文】本发明公开了一种油泥数控铣削装夹平台及夹装方法，明属于汽车油泥模型加工技术领域，用于解决现有的夹装平台找正麻烦、通用性差、精度低等问题，本发明的油泥数控铣削装夹平台，包含：T型槽平台、长凳子、短凳子、旋转轴、限位块、固定定位块、活动定位块和固定块；针对油泥铣削装夹而设计，充分利用T型槽的优势，使用定位块、活动定位块实现实现快速装夹找正，利用纵横T型槽实现坐标标定，从而无需预设坐标，利用长短凳子和矩阵孔的配合实现通用化。 【EN】The invention discloses an oil sludge numerical control milling clamping platform and a clamping method, which belong to the technical field of automobile oil sludge model processing and are used for solving the problems of troublesome alignment, poor universality, low precision and the like of the existing clamping platform, and the oil sludge numerical control milling clamping platform comprises: the device comprises a T-shaped groove platform, a long stool, a short stool, a rotating shaft, a limiting block, a fixed positioning block, a movable positioning block and a fixed block; the oil sludge milling fixture is designed for oil sludge milling, the advantages of the T-shaped grooves are fully utilized, rapid clamping and alignment are achieved by the aid of the positioning blocks and the movable positioning blocks, coordinate calibration is achieved by the aid of the longitudinal T-shaped grooves and the transverse T-shaped grooves, accordingly, coordinates do not need to be preset, and universalization is achieved by means of matching of long stools and short stools and matrix holes.

摘要：【中文】本发明公开了一种基于多尺度深度卷积神经网络的卫星图像目标检测方法，包括步骤收集卫星图像训练数据集，并进行样本标注；对卫星图像训练数据集进行预处理；搭建多尺度深度卷积神经网络；将预处理后的训练数据集输入到基于所述多尺度深度卷积神经网络的目标检测框架进行训练，获得训练好的目标检测神经网络；输入待检测卫星图像集，采用训练好的所述目标检测神经网络进行目标检测，输出识别结果。其显著效果是：提高了网络对于细粒度特征的检测结果以及区分不同物体的能力，改善了对于小物体和密集物体群的检测效果，具有更强的鲁棒性，有效地提高了目标检测效率，降低了硬件需求。 【EN】The invention discloses a satellite image target detection method based on a multi-scale depth convolution neural network, which comprises the steps of collecting a satellite image training data set and carrying out sample labeling; preprocessing a satellite image training data set; building a multi-scale deep convolution neural network; inputting the preprocessed training data set into a target detection framework based on the multi-scale deep convolution neural network for training to obtain a trained target detection neural network; inputting a satellite image set to be detected, adopting the trained target detection neural network to perform target detection, and outputting a recognition result. The remarkable effects are as follows: the method improves the capability of the network on detecting results of fine-grained features and distinguishing different objects, improves the detection effect on small objects and dense object groups, has stronger robustness, effectively improves the target detection efficiency, and reduces the hardware requirement.