11 results about "Orientation analysis" patented technology

The application discloses an AI identification and driving control system and method of a mobile inspection device, and relates to the technical field of driving control.The system comprises an image acquisition module, a biological identification and orientation analysis module, a same-direction pursuit judgment module, a driving direction execution module, and a continuous tracking and speed adjustment module;inspection images are acquired, target organisms are identified, and escape directions are predicted according to orientation information;whether a same-direction pursuit state is formed is determined by combining the moving direction of the device and the path area position;the driving direction is adjusted so that the included angle meets the threshold requirement and driving is performed;meanwhile, the position of the organism is continuously tracked;and collaborative control is realized by adjusting the speed and dynamically updating the driving direction.The application can effectively break the same-direction pursuit phenomenon, quickly drive the organism away from the inspection path, avoid the obstruction of the inspection process, and improve the continuous inspection operation efficiency and operation stability.

The application belongs to the technical field of production control, and provides a cone yarn automatic conveying control system and method for a beaming workshop. The method comprises the following steps: a vision device continuously shoots a plurality of groups of video image frames of a cone yarn, determines a sampling video image frame of each group of video image frames and a relative visual angle of the sampling video image frame and the vision device; a virtual three-dimensional model of the cone yarn is constructed based on the relative visual angle and the video image frame of the corresponding group, multi-dimensional features are extracted from each virtual three-dimensional model, multi-dimensional fusion feature sequences are obtained by performing feature splicing and fusion on each multi-dimensional feature, a feature fluctuation trend is obtained by performing secondary feature extraction on the multi-dimensional fusion feature sequences, and an orientation analysis result of the cone yarn is predicted based on the multi-dimensional fusion feature sequences and the feature fluctuation trend; if the orientation analysis result is inconsistent with a preset orientation, a correction device is triggered to adjust the orientation of the cone yarn. The application can realize accurate identification and correction of the direction of the cone yarn, improve the success rate of AGV grabbing, and further improve the beaming efficiency.

The application discloses an image orientation recognition method and device, equipment and a storage medium. The method comprises the following steps: using a projection neural network to preprocess a target image to obtain standard image data, the preprocessing comprising dimension reduction, noise reduction, normalization, cutting and gain control; encoding the standard image data according to APL neurons and a WTA mechanism through a preset KC layer to obtain encoded image data, the projection neural network and the preset KC layer comprising first synaptic weights subject to random distribution; inputting the encoded image data into a preset MBON layer for orientation analysis to obtain an orientation recognition result, the preset KC layer and the preset MBON layer comprising second synaptic weights updated based on a preset update algorithm. The application avoids the problems of catastrophic forgetting or low learning speed of the existing image orientation recognition model due to large distribution difference of samples with the same label, and avoids the problems of low learning and recognition efficiency of rare details in the image caused by the use of convolution kernels.

The invention relates to a picking robot and a fruit pose detection and interactive visualization method and system thereof, the picking robot comprises the fruit pose detection and interactive visualization system, the system adopts the following method to carry out fruit pose detection and visualization, and the method comprises the following steps: collecting an image of a fruit to be picked, and carrying out preprocessing; performing key point labeling on the to-be-picked fruit image based on the structure of the visible fruit calyx region to generate a training data set; setting a lightweight key point detection network, and performing fruit pose model training on the training data set based on the network; in the training process, the axial symmetry of a fruit calyx surface is utilized, the key point positioning precision is improved through axis alignment soft geometric constraint, and the geometric consistency and the anti-interference capability of a fruit pose model are enhanced; and based on the trained fruit pose model, an orientation analysis algorithm and a visual interface are adopted to realize visual display and interactive operation of pose information of to-be-picked fruits, and directly applied pose information is provided for the picking robot.

An information handling system peripheral camera analyzes captured visual images to detect a face and analyzes the face to determine an upright orientation. Based upon the face upright orientation, the camera defines a visual image orientation to communicate the visual image from the peripheral camera to an information handling system so that the face is communicated in an upright orientation for use in a video conference, such as with an indication of a landscape or portrait image orientation. In one embodiment, an information handling system can perform the face upright orientation analysis and command a camera orientation to the camera so that visual images of the camera are communicated in alignment with the face upright orientation.

A picking robot and a fruit pose detection and interaction visualization method and system thereof, the picking robot comprising a fruit pose detection and interaction visualization system, the system adopting the following method for fruit pose detection and visualization, the method comprising: collecting a picking fruit image and performing preprocessing; based on the structural nature of the visible calyx area, key point labeling is performed on the picking fruit image to generate a training data set; a lightweight key point detection network is set, and the fruit pose model is trained based on the network; in the training process, the axisymmetric nature of the fruit calyx surface is utilized to improve the key point positioning accuracy through axis alignment soft geometric constraints, and the geometric consistency and anti-interference ability of the fruit pose model are enhanced; based on the trained fruit pose model, an orientation analysis algorithm and a visualization interface are used to realize intuitive display and interactive operation of the pose information of the picking fruit, and direct application pose information is provided for the picking robot.

The present disclosure relates to a guide configured to allow movement of at least a portion of a light beam delivery system to adjust an orientation of the at least movable portion to different orientations of a patient's head measured about an axis of an eye to be treated. The system has an orientation analysis system for acquiring orientation related data using one or more body parts of the patient. The ophthalmic laser system allows a user and / or controller to move the movable portion of the light beam delivery system using the orientation related data such that during laser treatment, the movable portion has an orientation about the axis of the eye and relative to the patient's head that corresponds to or substantially corresponds to a predetermined target orientation; or the orientation is at or substantially within a predetermined target range.

The invention describes a method for detecting a closure (12; 12 ') applied to a body (4) of a respective package (2). 12 '). The detection device (20) comprises an imaging device configured to acquire at least one image (21) and an analysis unit; the analysis unit is configured to analyze each image (21) and to determine a position of the closure (12; 12 ') relative to a reference line (25) of the respective body (4). The analysis unit is configured to determine, from each image (21), at least one first reference point (26) of the at least one auxiliary portion (14, 14 ') of the respective closure (12, 12'), and to obtain a difference between the determined first reference point (26) and the at least one auxiliary portion (14, 14 ') of the closure (12, 12'). The analysis unit is further configured to determine an orientation angle of the straight line (27) with respect to the reference line (25), the orientation angle describing an orientation of the closure (12; 12 ') with respect to the body (4).

An active calibration method for a detection module, a system thereof and an electronic device. The active calibration method for the detection module comprises the steps of: performing a center position analysis on a first light spot image projected by a projection assembly via an industrial camera to correct a shift of a light source component and an optical component of the projection assembly; performing an orientation analysis on a second light spot image projected by the projection assembly via a receiving assembly to correct a rotation of the light source component and the optical component; performing a uniformity analysis on a third light spot image projected by the projection assembly via the industrial camera to correct a tilt of the light source component and the optical component; and performing a sharpness analysis on a fourth light spot image projected by the projection assembly via the industrial camera to correct a sharpness of the light source component and the optical component, so that the light source component and the optical component are positioned at a sharpest position.

The application discloses a kind of evaluation methods of the seepage capacity of compact rock layer, specifically related to material permeability characteristic measurement technical field, for solving the problem that the evaluation result is not accurate due to the anisotropy of seepage not being considered in the existing intelligent sensor layout;By first obtaining the seepage basic parameters of compact rock sample and the initial point of sensor layout, then using the seepage orientation analysis method to determine the potential orientation of the dominant seepage channel, and then coupling the anisotropy of fluid diffusion coefficient and the difference in pore throat tortuosity to determine the specific orientation, then adjust the layout posture and initial point of the intelligent sensor according to this orientation to align the probe with the dominant channel, finally control the sensor to collect signals and complete the evaluation;Make the sensor layout adapt to the anisotropy characteristics of rock stratum, can accurately capture the real signal of the dominant seepage channel, thereby significantly improving the accuracy and comprehensiveness of the evaluation result.

The application belongs to the field of high-temperature deformation and fracture, and discloses a crystal plasticity reverse modeling and high-throughput prediction method and system, which comprises the following steps: observing the morphology and phase change behavior of a sample at a first preset temperature in situ; cooling the heated sample to obtain a sub-phase organization sample; reconstructing the morphology and orientation information of the parent phase of the sample; constructing a crystal plasticity RVE model; fitting to obtain high-temperature constitutive model parameters; obtaining crystallographic and crystal deformation information of the sample; performing an SEM in-situ high-temperature tensile test on the sample; analyzing the stress and strain evolution in the deformation process of the sample in situ by using DIC technology; and verifying the prediction accuracy. The application completes the modeling of the high-temperature austenite parent phase organization based on the room-temperature sub-phase organization reverse, and realizes the orientation analysis, texture analysis, deformation morphology analysis, stress and strain analysis and the like in the high-temperature deformation process by combining with the crystal plasticity simulation, so as to realize the in-situ high-throughput characterization of the high-temperature deformation.