613 results about "Object distance" patented technology

The invention discloses a wiring terminal full appearance detection system and method based on an annular track platform, and the system comprises a track which is provided with a transparent material conveying chain plate segment and is used for the uniform motion transmission of a wiring terminal; the multi-station visual detection device comprises a plurality of visual camera groups and an object distance adjusting assembly, the plurality of visual camera groups are respectively arranged at different stations of the track and are used for carrying out image acquisition on six front-view main surfaces and a set surface of the wiring terminal, and the object distance adjusting assembly is connected with the visual camera groups and is used for carrying out image acquisition on the set surface of the wiring terminal. The object distance adjusting assembly is used for adjusting the object distance or / and the position of the linear scanning camera; the AI visual inspection module comprises an image marking unit used for carrying out defect classification and marking on the collected images; the model training unit is used for carrying out iterative training on the marked image based on a small sample learning algorithm to generate a multi-station detection model; and the real-time detection unit is used for deploying a multi-station detection model to execute component statistics, defect detection and defective product rejection.

The invention relates to a related light measuring device based on a bevel mirror and a control method, and solves the problem that the existing three-dimensional detection technology is insufficient in comprehensive adaptability and cannot meet the requirements of precision detection for stability, precision and data collaboration, and the related light measuring device comprises a laser emission head, a double-bevel mirror, a light path receiving module, a zoom camera and a circuit processing unit. The first bevel mirror is provided with a through hole for laser to directly irradiate a target; the light path receiving module receives and reflects the laser-to-electric signal; the second bevel mirror reflects the target and the light spot light to the camera to obtain visual information; and the circuit processing unit controls all the components, integrates laser and visual information, and realizes three-dimensional detection of the target. The method has the following effects: through the light path collaborative design of the double-bevel mirror, precise and synchronous acquisition of laser ranging signals and visual contour / light spot information is realized, so that high-precision three-dimensional detection of the distance, the flatness and the segment difference of a target object is efficiently completed.

The present invention provides an automatic focusing method, an electronic device, and a readable storage medium. The automatic focusing method comprises: using a target detection model to detect a target object in an image frame acquired at an initial focal length by an optical imaging device to be focused, so as to acquire a target object detection result; on the basis of the target object detection result, determining whether a target object is present in the image frame; if yes, determining an actual object distance on the basis of the target object detection result and the initial focal length, and on the basis of the actual object distance and a mapping relationship between the object distance and an optimal imaging focal length, determining the optimal imaging focal length; and if not, using a preset search algorithm to search for a focal length until an image having the highest definition value is found, and using a focal length corresponding to the image having the highest definition value as the optimal imaging focal length. The present invention primarily employs a deep learning-based automatic focusing method, supplemented by an image definition evaluation method. Compared with traditional passive focusing methods, the present invention greatly improves the focusing efficiency. Compared with traditional active focusing methods, the present invention eliminates a need for adding an additional optical ranging component, ensuring that the imaging device has a simple structure and a low cost.

A variable magnification optical system comprising a first lens group, a second lens group, a third lens group, and a rear group in order from object side is configured so that at varying magnification the distances between adjacent lens groups are varied; at a predetermined object distance the variable magnification optical system has focused states with different amounts of aberration; the rear group includes a first focusing lens group and a second focusing lens group disposed closer to an image side than the first focusing lens group, the first and second focusing lens groups moving along different trajectories at focusing; at the object distance the first and second focusing lens groups move at transition from a first focused state to a second focused state; and the following conditional expression is satisfied:−6.80<f1 / f2<−0.05where f1 and f2 are the focal lengths of the first and second lens groups, respectively.

The invention relates to the technical field of teleoperation, in particular to a teleoperation robot auxiliary control method based on a safety margin theory. The method comprises the following steps: establishing a three-dimensional digital twinborn model according to an image acquisition device, and obtaining a target object distance, a communication time delay size and a motion tracking error; based on the three factors, a fuzzy control theory is adopted to calculate a teleoperation safety margin value, and the safety of the current teleoperation is quantitatively evaluated. And calculating a master-slave motion mapping ratio and an end view display ratio according to the safety margin value. And constructing virtual force guidance according to the safety margin value, feeding back the virtual force guidance to an operator through force feedback equipment, and guiding the operator to complete a teleoperation task in combination with visual interaction equipment. Compared with a traditional teleoperation method, the method can effectively improve teleoperation efficiency and safety, and can better adapt to a complex long-distance working environment.

The invention discloses a real-time automatic focusing hyperspectral acquisition method and a real-time automatic focusing hyperspectral acquisition device. The real-time automatic focusing hyperspectral acquisition method comprises the following steps: step 1, acquiring dark background pixel signals; 2, automatic focusing is carried out; 3, repeating the step 2 until the requirement is met, recording automatic focusing data, and performing spectrum acquisition to obtain original spectrum data; step 4, calculating a full-width object distance matrix of the original spectral data based on the phase difference finally obtained in the step 3; 5, correcting the original spectral data based on the full-width object distance matrix to obtain corrected spectral data; and step 6, calculating the corrected reference plate spectral data based on the corrected spectral data and the dark background pixel signal, calculating the reflectivity of the target object to be measured, and realizing real-time automatic focusing hyperspectral acquisition. The method innovates a phase difference analysis method, and efficiently processes light change and target offset in a dynamic scene; the combination of the two improves the focusing precision and stability, effectively solves the problem that the scanning target in the dynamic scene is liable to defocus, and achieves the efficient adaptation of the dynamic scene.

The read-write table device comprises a shell, a spectroscope, an eyepiece and an object distance adjusting panel, the shell is of a closed frame structure with an opening in the front side wall, the eyepiece is embedded in the rear side wall of the shell, and the eyepiece comprises a concave mirror and a retroreflection lens tightly surrounding the peripheral side of the concave mirror; a spectroscope is obliquely arranged in the shell between the opening of the shell and the eyepiece, an object distance adjusting panel is embedded in the lower side wall of the shell, and the size of the object distance adjusting panel is matched with that of the retroreflection lens, so that the middle area of a picture seen by human eyes through the read-write table device is a virtual image, the peripheral area is a real image, and the myopia prevention and control effect can be achieved. The included angle between the side wall of the shell and the eyepiece meets a certain relational expression, so that the included angle between the side wall of the shell and the eyepiece can be determined according to the emergent angle of the edge light, the viewing effect that the visual field is not shielded by the shell is achieved, and the use experience and the visual effect are improved.

The invention relates to a read-write table for realizing peripheral myopia type defocus, which comprises a totally-closed hollow rectangular shell, a spectroscope, a concave mirror, a retroreflection multi-structure combined mirror and an object distance matching light-transmitting panel, and is characterized in that the spectroscope is obliquely arranged in the middle of the shell, the concave mirror is embedded into the rear side wall of the shell, and the front side wall of the shell corresponding to the concave mirror is open; a retroreflection multi-structure combined mirror is embedded into the upper side wall of the shell, the retroreflection multi-structure combined mirror comprises a central hollow area I and a combined mirror area I, and the combined mirror area I comprises retroreflection lenses and concave lenses which are alternately arranged from inside to outside; an object distance matching light-transmitting panel is embedded into the lower side wall of the shell, the object distance matching light-transmitting panel comprises a second central hollow area and a second combined lens area, and the second combined lens area comprises convex lenses and plane lenses which are alternately arranged from inside to outside; according to the invention, the middle area of a picture seen by the human eyes through the read-write table is a virtual image, and virtual images and real images in the peripheral area are alternately distributed, so that the myopia prevention and control effect is realized, and the visual continuity and definition are ensured.

The invention belongs to the field of image spectrum detection, and particularly relates to a high-precision detection device and method for a liquid meniscus contour in a capillary tube. Comprising a telecentric lens, an industrial camera, a displacement platform, a light source, a capillary tube and a culture dish, the industrial camera, the capillary tube and the culture dish are installed on different displacement platforms respectively, the light source is provided with a rotary telescopic rod and can move in all directions, when the liquid meniscus contour in the capillary tube is shot, the light source, the meniscus of the capillary tube and the industrial camera are adjusted to be aligned to the same horizontal plane, and on the basis of the telecentric imaging principle, according to the light refraction principle, the liquid meniscus contour in the capillary tube is obtained. And deriving a distortion correction function of the meniscus contour in the capillary tube. According to the invention, the industrial camera and the telecentric lens are matched for use, the problem of shape distortion caused by object distance change is avoided, and the difficulty of image distortion correction is reduced. The meniscus contour distortion correction function derived based on the telecentric imaging principle can accurately calculate the actual value of the meniscus contour and restore the actual meniscus contour.

The invention discloses a dual-light registration and infrared camera calibration method, which comprises the following steps of: synchronously acquiring a visible light image and an infrared image, and acquiring multiple groups of dual-light image pairs under different distances within a preset object distance range; cross-modal feature key points are extracted, a matching corresponding relation is established, registration parameters under object distances are generated, and a dynamic compensation database is constructed; lead arrays are embedded in a plurality of horizontal edges and a plurality of vertical edges of a traditional checkerboard calibration plate, and leads are driven to generate uniform heat radiation; checkerboard corner coordinates are accurately extracted based on developed heating side lines in the infrared image, a real-time target detection model software deep learning model is utilized to construct a multi-scale feature fusion network, and combined geometric fitting is carried out on checkerboards in the bimodal image; and carrying out adaptive region segmentation on the fitted checkerboard, and reconstructing a black and white alternate state of the visible light checkerboard. The technical problem of registration failure in a complex environment can be solved, and the stability and repeatability of calibration are improved.

The invention relates to the technical field of tunnel crack detection, and discloses a tunnel crack image identification method and a tunnel wall uniform image detection system. The method comprises the steps that tunnel images are collected, images belonging to cracks are calibrated to form an image set, and a crack target sample library is constructed; performing target detection identification on the collected tunnel image set, optimizing a target identification method according to an identification result, and outputting multiple tunnel crack disease detection images; extracting crack features from the collected tunnel image set; and based on the multiple tunnel crack disease detection images, performing pixel-level segmentation on the tunnel crack disease to complete tunnel crack identification. According to the invention, the traditional layout is broken, the requirements of closer shooting object distance and depth of field are met, and the detection requirement of smaller width of the welding seam is met; according to the algorithm level, a front image preprocessing part of deep learning training is optimized, especially contrast enhancement, multi-level fusion and the like are added, and the recognition accuracy of a tiny target can be further improved.

The invention discloses an industrial lens. The industrial lens comprises a first lens group, a second lens group and a third lens group, the positions of the first lens group and the third lens group are fixed, and the position of the second lens group is floatable; the first lens group comprises a first lens, a second lens, a third lens, a fourth lens, a fifth lens and a sixth lens; the second lens group comprises a seventh lens, an eighth lens, a ninth lens, a tenth lens, an eleventh lens and a twelfth lens; the third lens group comprises a thirteenth lens and a fourteenth lens. The second lens group moves along the optical axis to realize focusing of different object distances, and the first lens group and the third lens group are fixed, so that the influence of focusing to different object distances on resolution can be reduced, and the balance of image quality of each working distance is ensured. And the number of the lenses in each lens group is reasonably set, so that the industrial lens with high image quality imaging performance can be realized.

According to the multi-object-distance and low-distortion lens for machine vision and the camera device, the lens comprises an optical system, the optical system comprises a first lens group, an aperture diaphragm and a second lens group, the first lens group comprises a first lens, a second lens and a third lens, and the second lens comprises a second lens and a third lens. The second lens group comprises a fourth lens, a fifth lens, a sixth lens, a seventh lens and an eighth lens which are arranged on the same optical axis from the object space to the image space; the first lens is a biconvex lens with positive focal power, the second lens is a biconvex lens with positive focal power, the third lens is a biconcave lens with negative focal power, the fourth lens is a biconvex lens with positive focal power, the fifth lens is a biconcave lens with negative focal power, the sixth lens is a meniscus lens with negative focal power, and the seventh lens is a meniscus lens with positive focal power. And the eighth lens is a biconvex lens with positive focal power. The lens provided by the invention has the advantages of large object distance range, low distortion and high resolution.

The invention relates to a dynamic tumor tracking method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring 4DCT image data of a patient, and performing mixed data enhancement on the 4DCT image data to generate a 3DCT image representing each respiratory stage of the lung; based on the 3D CT image, generating an X-ray image in a respiratory cycle of a simulated patient by adopting a mathematical simulation method; registering the X-ray image and the 4DCT image data to generate a tumor tracking model of the patient; and inputting an X-ray image of a patient in a radiotherapy process into the tumor tracking model, and outputting a tumor tracking result through the tumor tracking model. According to the invention, real-time three-dimensional tracking of the tumor can be realized by using a single low-dose X-ray image under any geometric condition, high-precision tumor positioning can be provided on the premise of ensuring low radiation dose, the method is suitable for any projection angle and source-object distance change conditions, and various geometric relationship changes in the treatment process can be flexibly coped with.