26 results about "Visual estimation" patented technology

The invention discloses a vehicle-road collaborative perception and fusion estimation vehicle stability control method for a beyond-visual-range scene, and belongs to the technical field of crossing of intelligent traffic, vehicle control engineering and an automobile safety technology. The method comprises the following steps: acquiring vehicle dynamic data through a vehicle end sensor, estimating a dynamic pavement adhesion coefficient based on a seven-degree-of-freedom model and unscented Kalman filtering, acquiring a pavement image by using a roadside camera, identifying a pavement state by using a MobileNetV3 + model, and mapping the pavement state into a visual pavement adhesion coefficient; a discrete time Kalman filter is adopted to fuse dynamics and visual estimation results, and the fusion weight is dynamically adjusted according to the confidence coefficient; and transmitting the fused road adhesion coefficient to a vehicle end controller, and designing an upper-layer controller and a lower-layer controller based on sliding mode control and road adhesion coefficient identification, thereby realizing stability control of the vehicle under complex road conditions. The accuracy and foresight of road adhesion coefficient estimation can be effectively improved, and the handling stability of the vehicle under the limit working condition is enhanced.

The invention discloses a visual estimation method and device for an included angle of a hinge headstock, relates to the technical field of computer vision and intelligent perception and measurement of vehicles, and can solve the problems of high cost, poor environmental adaptability, insufficient robustness and lack of end-to-end direct prediction capability during measurement of the included angle of the headstock in the prior art. Comprising the following steps: synchronously acquiring image data and head-mounted included angle truth value data of a trailer, and performing time alignment processing on the image data and the head-mounted included angle truth value data to form a training data set; an AngleNet end-to-end head-hanging included angle estimation model is constructed, the estimation model comprises an input layer, a backbone network, a regression head and a loss function, the backbone network is a progressive deformable convolutional network GDNet, and the loss function adopts TMSE Loss; training the estimation model by using the training data set; and deploying the trained estimation model to a vehicle-mounted computing platform, processing an image acquired by the camera in real time, and outputting a head-mounted included angle estimation value. The method is used for hinge headstock hanging included angle visual estimation.

The utility model relates to a packing machine lining paper negative pressure disc adjusting device, and belongs to the technical field of cigarette production, the packing machine lining paper negative pressure disc adjusting device comprises a calibration device and a positioning round block, the calibration device comprises a calibration plate, a sliding support and a locking wheel sleeve, the sliding support and the locking wheel sleeve are arranged at the two ends of the calibration plate, and the sliding support is provided with an ejector pin positioned with a negative pressure disc air suction hole; through the sliding supports with the ejector pins arranged on the calibration plate, the ejector pin on the sliding support at one end is in contact with the negative pressure disc air suction hole in the right side, the other end is in contact with the negative pressure disc air suction hole in the middle, the problem that the position is judged through visual inspection of a steel ruler is solved, and repeated adjustment of the cog belt is not needed. The position synchronization of the air suction holes of the middle negative pressure disc and the air suction holes of the negative pressure discs on the two sides is guaranteed, the calibration accuracy is guaranteed, the calibration difficulty is greatly reduced, meanwhile, the problem that the eccentric disc is disassembled and reassembled after being adjusted on the negative pressure cutting device is solved, and the calibration efficiency is improved.

The invention provides a method and a system for quickly positioning railway intrusion personnel based on visual estimation, belongs to the technical field of target identification, applies an advanced monocular depth estimation and positioning method to rail transit, and solves the potential safety hazard of train operation caused by insufficient personnel intrusion position along the current railway. According to a pre-constructed target prior mapping relation between image pixel coordinates and actual geographic coordinates, after an invader is detected, the pixel coordinates of the invader are directly mapped to corresponding actual longitude and latitude positions; and if the target prior mapping information does not exist, calling a coordinate conversion algorithm according to the pixel coordinates to realize conversion from the pixel coordinates to latitude and longitude coordinates, realizing dynamic conversion from an image plane to a global geographic coordinate system, and obtaining actual latitude and longitude positioning of personnel invasion. According to the invention, high-precision positioning can be carried out on the invading personnel in the rail transit scene only by means of the monocular camera, so that the problem that personnel invasion is difficult to check in an existing system is solved, and the safe operation level of a railway is improved.

The invention provides an image assisting method and system, an electronic device, a storage medium and a program product, and belongs to the technical field of intelligent medicines.The method comprises the steps that an operation image is obtained based on a camera device on a hot steam ablation operation catheter; based on the parameters of the camera equipment, generating image coordinates and superposing the image coordinates on the surgical image; based on the image coordinates, determining a sheath tube moving distance of the hot steam ablation operation; under the condition that the movement distance of the sheathing canal reaches the target distance, a surgical operation prompt is initiated. According to the method, the camera device on the hot steam ablation operation catheter is used for obtaining the operation image, additional endoscope equipment is not needed, the sheath tube moving distance of the hot steam ablation operation is determined through the image coordinates overlapped on the operation image, and under the condition that the sheath tube moving distance reaches the target distance, the operation prompt is initiated; the defects that judgment of a doctor on the moving distance of the sheathing canal completely depends on visual estimation and operation hand feeling, standard operation is difficult to achieve, and the uniformity of the operation effect is affected are overcome.

This invention discloses a distributed multimodal UAV angle perception and prediction method, comprising the following steps: S1, determining the system model; S2, determining the visual signal and radio frequency signal models; S3, fusing visual information from multiple cameras to estimate the UAV's orientation relative to the base station at the current moment; S4, fusing the visual estimation results from historical time slots with the wireless echo perception results to predict the UAV's state at the next moment; S5, based on the UAV's state at the next moment predicted in step S4, calculating and generating the optimal transmission beam for the base station antenna in advance. This invention eliminates the blind spots of a single base station's perception through a distributed camera network, achieving omnidirectional, continuous, and robust positioning of UAVs in complex urban environments; by fusing visual and wireless echo information for high-precision motion prediction, the communication system can pre-align with the UAV, transforming passive tracking into proactive service, thereby significantly improving the stability and transmission efficiency of low-altitude communication links.

The invention discloses a spray gun position control method, device and equipment, a storage medium and a forcible entry robot, and belongs to the technical field of building equipment. Processing the gap detection data by adopting a quartile distance method to obtain effective gap data, and calculating an average slope based on the effective gap data; and based on the average slope, the critical position of the to-be-forcible-entry structure is calculated, and a spray gun of the forcible-entry robot is controlled to work based on the critical position. The method effectively replaces a traditional mode depending on manual visual estimation, so that the problem that the initial position of the spray gun is improper due to manual judgment errors is directly avoided, and unification of construction safety and operation efficiency is achieved.

The invention relates to the technical field of vehicle control, in particular to a vehicle control method and system based on the head posture of a driver, and the method comprises the following steps: obtaining first inertial data of the head of the driver, and synchronously collecting image data containing optical mark points on a head-mounted device; based on the image data, calculating to obtain a visual estimation posture of the head of the driver; based on the first inertial data, calculating to obtain an inertial estimation posture of the head of the driver; performing data fusion on the visual estimation attitude and the inertial estimation attitude to obtain six-degree-of-freedom attitude information of the target; acquiring head posture information and a driving instruction of a driver, and establishing a mapping rule between the head posture information and the driving instruction information; and according to a mapping rule, converting the target six-degree-of-freedom attitude information into a corresponding vehicle control instruction, and sending the vehicle control instruction to a vehicle execution mechanism to control vehicle actions, so that the vehicle driving safety of a driver can be improved.

In order to help pilots relying on manual visual inspection to realize autonomous landing more safely, the invention provides a method for assisting autonomous landing by using an airborne Doppler navigation radar and an airborne radar altimeter. Firstly, the glide speed can be obtained by directly utilizing the existing speed detection data of the airborne Doppler navigation radar. And secondly, calculating an inclination angle in a vertical plane by using detection data of Doppler frequency shift by a Doppler navigation radar, thereby obtaining a glide angle. Or the glide angle is calculated according to the trigonometric function relation by using the three-dimensional velocity component measured by the airborne Doppler navigation radar. And then on the basis of acquiring height measurement data by using an airborne radar altimeter, calculating a glide distance by using the acquired glide angle and a trigonometric function relationship. Furthermore, if the aircraft is further provided with a flight management system, an automatic driving system and the like, technical support can be provided for semi-automatic landing by using landing data provided by an airborne Doppler navigation radar and a radar altimeter.

Method for quantifying an impression of volume given by a head of hair (4), comprising the steps consisting in: a) acquiring a front image and / or a side image of the head of hair of a person (1) to be evaluated, b) segmenting the head of hair in each image, c) defining, within the segmented head of hair in the front image, at least one region of interest comprising hair roots (7) and / or at least one region of interest comprising hair lengths (8); and / or defining, within the segmented head of hair in the side image, a region of interest comprising a mass of hair (6), d) delivering a value of at least one characteristic parameter, extracted from one of the regions of interest or corresponding to the sum of the intermediate parameters extracted from several regions of interest, the characteristic parameter being associated with a visual estimate v of the volume given by the head of hair in the regions of interest.

The invention provides a trailer hinge angle real-time estimation method and system based on feature point tracking. The method comprises the steps of obtaining a first image frame of a target trailer at a current moment and a second image frame of the target trailer at a previous moment; determining a first feature point list of the first image frame according to a second feature point list in the second image frame; according to the first feature point list and a preset feature point template, calculating to obtain a visual estimation angle and confidence; a first Kalman filter at the current moment is updated according to the visual estimation angle and the confidence coefficient, an updated second Kalman filter is obtained, and a hinge angle estimation value at the current moment is output based on the second Kalman filter; and within a preset time period after the current time, according to the motion data of the target trailer, outputting the hinged angle estimated value of each time within the preset time period through the second Kalman filter, thereby improving the accuracy and real-time performance of real-time estimation of the hinged angle of the trailer.

The application discloses a kind of amoxicillin solubility visual detection methods based on quantitative color scale, specifically related to drug detection field, the present application aims at solving the problems of large standard gradient interval, unstable color development and lack of correction in the existing visual comparison method, method includes: with 0.2M potassium dihydrogen phosphate solution as solvent, preparation amoxicillin saturated solution;Linear regression equation is established, by precisely preparing 9 concentration gradient amoxicillin standard solution, after adding copper sulfate solution, form stable yellow-green turbidity gradient, and draw linear regression equation with absorbance value;Sample determination, the sample liquid is compared with standard gradient and turbidimetric analysis is carried out, the concentration is determined, and the solubility is calculated by combining dilution multiple, the present application converts visual estimation into accurate quantification by high-density gradient standard and linear regression model, significantly reduces matching error, improves anti-interference ability and accuracy, and is suitable for rapid detection in drug research and development and quality control.

The invention relates to the technical field of image recognition, in particular to a saline-alkali soil bacteria and algae coverage degree detection method based on machine vision, which comprises the following steps: according to a collected saline-alkali soil surface image, setting a sliding pixel window with a fixed size to traverse saline-alkali crust and biological crust areas in the image; joint probability distribution of pixel gray values is calculated in four directions in each sliding window, and an omni-directional gray co-occurrence matrix set is obtained. According to the method, internal cavities of the connected areas are filled and the edges are smoothed in cooperation with morphological closed operation, the segmentation breakage phenomenon caused by uneven internal textures is eliminated, it is ensured that the generated distribution mask is highly attached to the actual biological crust physical boundary, and the specific value of pixel-level accumulative statistics and the total amount of the total image is calculated based on the final mask. Deviation of artificial subjective visual estimation is eliminated, and reliable data dimensions are provided for ecological restoration monitoring.

This invention relates to a sensor fusion indoor positioning method for an unmanned ground vehicle, comprising the following steps: 1. Establishing a discretized kinematic model of a four-wheel drive vehicle; 2. Establishing a visual measurement model based on camera and inertial measurement unit (IMU) measurement data; 3. Designing a model based on the visual measurement model and an extended H... ∞ 4. Design a visual estimator for the filter; 5. Design a lidar estimator based on the adaptive Monte Carlo localization method; 6. Design a fusion estimator based on the visual estimator and lidar estimator; The method of this invention is based on the establishment of a visual estimator, a lidar estimator and a fusion estimator, which realizes the robustness of indoor localization for unmanned ground vehicles and solves the problems of camera information loss, noise model uncertainty and poor localization reliability of lidar in similar map scenes in the prior art.

Quantification of an impression of volume conferred by hair Method for quantifying an impression of volume conferred by hair (4), comprising the steps of: a) acquiring a front image and / or a profile image of the hair of a person (1) to be evaluated, b) segmenting the hair on each image, c) defining within the hair segmented on the front image at least one region of interest comprising hair roots (7) and / or at least one region of interest comprising hair lengths (8);and / or define within the segmented hair on the profile image a region of interest comprising a mass of hair (6), d) deliver a value of at least one characteristic parameter, extracted from one of the regions of interest or corresponding to the sum of intermediate parameters extracted from several regions of interest, the characteristic parameter being associated with a visual estimate v of the volume conferred by the hair in the regions of interest. Figure for the abstract: Fig. 1;

This invention relates to a method for detecting the width of rice grains. The method utilizes a rice grain width detection device, which includes a panel with a fixed width baffle, a fixed length baffle, and ten movable width baffles. The movable width baffles are parallel to the fixed width baffle and distributed along the extension direction of the fixed length baffles. A movable width baffle state-maintaining mechanism is provided between the movable width baffles and the panel. The movable width baffle closest to the fixed width baffle, the fixed length baffle, and the fixed width baffle together form a first rice grain storage blind slot. Two adjacent movable width baffles and fixed length baffles form a second rice grain storage blind slot. This invention has the advantage of accurately detecting the width of rice grains, solving the problem of low detection accuracy caused by manually aligning rice grains side-by-side by hand for visual estimation.

The invention discloses an amoxicillin solubility visual detection method based on quantitative color gradation, particularly relates to the field of drug detection, and aims to solve the problems of large standard gradient interval, unstable color development and lack of contrast correction in the existing visual comparison method. Preparing an amoxicillin saturated solution; the method comprises the following steps: establishing a linear regression equation, precisely preparing an amoxicillin standard solution with 9 concentration gradients, adding a copper sulfate solution to form a stable yellow-green turbidity gradient, and drawing the linear regression equation according to an absorbance value; when the sample is measured, turbidimetric analysis is carried out on the to-be-measured liquid and the standard gradient, the solubility is calculated in combination with the dilution multiple after the concentration is determined, visual estimation and reading are converted into accurate quantification through the high-density gradient standard and the linear regression model, the matching error is remarkably reduced, and the anti-interference capability and accuracy are improved; the method is suitable for rapid detection in drug research and development and quality control.

The invention discloses an orchard crop yield monitoring robot and a control method thereof, and relates to the field of agricultural robots. The robot is composed of a yield monitoring inspection system and a crawler-type chassis walking system. According to the control method disclosed by the invention, the grade of the fruit maturity is determined based on the color characteristics of the fruit and the pre-estimated volume through a visible fruit maturity estimation method, so that the extraction precision of the fruit in a visual field range under natural illumination and complex backgrounds is remarkably improved; the total theoretical fruit yield is determined based on the shielding compensation coefficient and the fruit visual estimation yield through a shielding fruit compensation yield correction method, and the problem of shielding compensation in the canopy is solved; a complete panoramic image covering the vertical direction of the fruit tree canopy is formed through the vertical panoramic splicing method of the fruit tree canopy, so that the integrity of monitoring data is ensured. According to the invention, the problems of accurate monitoring of crop yield and shielding correction in a complex orchard environment are solved, agricultural digital yield estimation is assisted, the labor cost is reduced, and the economic benefit is improved.

This invention provides a regional wave measurement system and method based on a wave glider and monocular vision, belonging to the field of ocean wave observation technology. The method includes the following steps: installing an airborne sensor, an airborne computing unit, a monocular camera, and a single-point wave measurement unit on a wave glider platform; calculating single-point reference wave parameters; performing monocular vision-based sea surface image preprocessing and feature analysis; based on the single-point reference wave parameters, calibrating and scaling the amplitude and spectrum of the local sea surface wave field image information; fusing the single-point wave measurement and visual estimation results; and calculating the wave height, wave direction, wave period, and directional spectrum within the fused local sea surface wave field. This invention, based on a wave glider platform, combines high-precision single-point wave measurement with sea surface images acquired through monocular vision to achieve high-precision measurement of wave parameters within a certain area around the wave glider.

The application discloses a kind of hinge truck head hanging clip angle visual estimation method and device, it is related to computer vision and vehicle intelligent perception and measurement technical field, can solve the problems of high cost, poor environmental adaptability, insufficient robustness and lack of end-to-end direct prediction capability when existing technology head hanging clip angle measurement.Therein, the image data and head hanging clip angle true value data of trailer are synchronously collected, and the image data and head hanging clip angle true value data are time-aligned to form a training data set;An AngleNet end-to-end head hanging clip angle estimation model is constructed, the estimation model includes an input layer, a backbone network, a regression head and a loss function, wherein the backbone network is a progressive deformable convolution network GDNet, and the loss function uses TMSE Loss;The training data set is used to train the estimation model;The trained estimation model is deployed to a vehicle-mounted computing platform, real-time processing of the image collected by the camera is carried out, and a head hanging clip angle estimation value is output.The application is used for hinge truck head hanging clip angle visual estimation.

The invention discloses an unloading vehicle container angle measurement method, an unmanned dumper and a driving control method thereof, and the measurement method comprises the steps: collecting an image through a side rearview mirror camera of an unloading vehicle, and obtaining a current rearview mirror image; identifying a region of interest of the current rearview mirror image by using a pre-trained image segmentation model to obtain a current position feature value corresponding to the region of interest; a pre-established discrete mapping table is called, the discrete mapping table is configured with multiple sets of position feature values and cargo box state values which are in one-to-one correspondence, and the cargo box state values are numerical values quantized according to the cargo box lifting degree; determining a current container state value matched with the current position characteristic value according to a discrete mapping table; and according to the container initial inclination angle and the container inclinable limit angle of the unloading vehicle, the current container state value is converted into the container current inclination angle. A camera is used for capturing an image of a container from a vehicle-mounted view angle, and the container lifting angle is visually estimated based on a deep learning algorithm model.

The invention relates to a system and method for estimating global rig state. The system comprises a model incorporating multiple variables related to rig state, at least one camera operably connected to at least one processor wherein said camera is capable of gathering visual data regarding at least one variable of rig state and said processor is capable of compiling rig state data, estimating global rig state, or both. The system further comprises multiple sensors for measuring variables related to global rig state wherein said sensors are operably connected to said processor. The method comprises sensing various aspects of the rig state, collecting visual data corresponding with said sensor data, compiling multiple sources of rig data, and estimating the overall rig state.

This invention presents a vehicle stability control method for vehicle-road cooperative perception and fusion estimation in beyond-line-of-sight scenarios, belonging to the interdisciplinary fields of intelligent transportation, vehicle control engineering, and automotive safety technology. It acquires vehicle dynamics data from vehicle-side sensors and estimates the dynamic road surface adhesion coefficient based on a seven-degree-of-freedom model and unscented Kalman filtering. After acquiring road surface images using roadside cameras, the MobileNetV3+ model identifies the road surface state and maps it to the visual road surface adhesion coefficient. A discrete-time Kalman filter is used to fuse the dynamic and visual estimation results, dynamically adjusting the fusion weights based on confidence levels. The fused road surface adhesion coefficient is transmitted to the vehicle-side controller. Upper and lower level controllers are designed based on sliding mode control and road surface adhesion coefficient recognition to achieve vehicle stability control under complex road conditions. This method effectively improves the accuracy and predictability of road surface adhesion coefficient estimation, enhancing vehicle handling stability under extreme conditions.

The invention discloses a dish picking identification method and system based on visual estimation, and belongs to the field of target identification. According to the method, an image sequence is obtained through a monocular camera, and target identification (obtaining types and positions of tableware and dishes) and monocular depth estimation (obtaining physical depth information) are respectively carried out by using a pre-training model; then, carrying out comprehensive judgment by combining the space overlapping degree (such as intersection-to-union ratio) of the positions of the tableware and the dishes and the depth difference between the tableware and the dishes; when the overlapping degree is larger than the threshold value and the depth difference is smaller than the threshold value, it is judged that a dish clamping event occurs, and the dish category is output to trigger subsequent intelligent services (such as playing a baking video of the corresponding dish). According to the method, the existing camera of the catering store is used for completing food clamping action recognition when customers have meals, detection errors caused by lack of three-dimensional information are reduced, and a foundation is laid for realizing accurate service of the store.

The invention provides a single-target vision-assisted Beidou observation station coordinate monitoring method, which comprises the following steps of: firstly, acquiring initial coordinates of an observation station and a target, and performing coordinate conversion to obtain initial east-north-sky coordinates of the observation station; in real-time monitoring, the current pixel coordinates of the target and the current Beidou coordinates of the observation station are collected, the visual displacement is obtained by comparing the current pixel coordinates and the initial pixel coordinates of the target, northeast coordinate correction is carried out by combining a preset visual azimuth angle, and the visual northeast displacement of the target is obtained. And then performing visual estimation based on the displacement and the initial east-north-sky coordinates of the observation station to obtain the current visual east-north-sky coordinates of the observation station. And finally, fusing with the current Beidou coordinate of the observation station, and outputting the current northeast-east-sky coordinate of the observation station. According to the single-target vision-assisted Beidou observation station coordinate monitoring method provided by the invention, Beidou positioning is assisted based on single-target vision, the monitoring precision and continuity in an environment in which Beidou signals are shielded or interfered are improved, and stable and reliable displacement monitoring capability is maintained in a complex environment.

The invention relates to the field of hoisting, and discloses an adjustable balance hoisting beam and a hoisting system.The adjustable balance hoisting beam comprises a beam body; the lifting frames are respectively arranged at the two end parts of the beam body and are connected with lifting appliances for connecting lifting objects; the auxiliary positioning light-emitting device is arranged at the central position of the beam body, and the auxiliary positioning light-emitting device is arranged in a way that when the beam body extends in the horizontal direction, the auxiliary positioning light-emitting device can vertically and downwards emit light so as to be aligned with the gravity center of the hoisted object hoisted by the hoisting tool. By means of the adjustable balance hoisting beam, an operator can rapidly and visually align a lifting hook to the actual gravity center of a hoisted object in the sight blocking scene, the aloft work scene or the precise equipment hoisting scene where the gravity center needs to be accurately aligned, and compared with a method for estimating the gravity center of the hoisted object through visual inspection, subjective errors caused by manual judgment can be overcome; therefore, the hoisting precision and the operation safety are guaranteed.