62results about How to "Reduce the impact of light" patented technology

The invention includes the hardware devices composed of the pick-up head, the image interface and the processor, the time periodic image collection, the recognizing the direction of the person's eye gazing at as well as the non-contact intelligent control technique. In the said non-contact intelligent control technique, the direction of the person's eye gazing at is used as the control command or the auxiliary information. The invention can be applicable to home appliances, game machines, curing and nursing instruments or as the non-contact control devices for handicapped people etc. The gazing method can turn on or off some simple appliances such as air conditioners, 'waken' the robot or computer in the standby state. The gazing method also can indicate the user's intended control command.

The invention discloses an automatic guidance system based on radio frequency identification labels and vision, and the method of the automatic guidance system. The invention is characterized in adopting a four-wheel structure, wherein two front wheels are used for steering and two rear wheels are used for driving; a direct current motor is used as the travelling and driving device, which is driven to steer by a stepping motor; the invention is provided with a wheel steering angle positioning device, so the invention has relatively high dynamic response ability; a black and white parallel guide belt is adopted as a guide path, with the radio frequency identification labels discontinuously laid below, and red work station characters arranged at both sides; CCD collects road surface information, and the brightness of light source adaptive controls according to image gray information; characters are extracted by utilizing color differences; straight lines are rapidly HOUGH transformed and identified; a control signal is sent to AGV controller after processed by an industrial control computer, and the direct current motor and the stepping motor are controlled. In order to effectively realize the communication dispatch between AGVS in real time, radio frequency identification labels are adopted for positioning, and a wireless transceiver module is adopted for communication between AGVS and a host. The invention has the advantages of good environmental adaptability and navigation precision, and lower cost.

The invention discloses a robot intelligent obstacle avoidance system and method based on stereoscopic vision. The system comprises a binocular camera, an industrial personal computer and an actuator.The industrial personal computer is connected with the binocular camera and the actuator respectively. The method comprises the steps that the binocular camera acquires video image information in theworking area of a robot, and sends the video image information to the industrial personal computer; the industrial personal computer receives video image information sent by the binocular camera, andcalculates the depth of each pixel according to the video image information and the internal and external parameters of the binocular camera to generate a depth image in the working area; the industrial personal computer further identifies an obstacle in the depth image according to deep learning, carries out three-dimensional reconstruction on the obstacle according to the depth image, calculates the three-dimensional coordinates of the obstacle, compares the three-dimensional coordinates with the motion path of the robot and detects whether the robot collides with the obstacle, adjusts themotion paths of the robot, generates corresponding motion path information, and transmits the motion path information to the actuator; and the actuator receives the motion path information given out by the industrial personal computer, and adjusts the motion path of the robot.

The invention belongs to the fields of fruit tree cultivation and fruit tree ecology, and discloses a shaping method for a downward-arrowhead-shaped tree shape of a pear tree. The tree shape has a tree height of 2.4-2.5 m, and a trunk height of 0.6-0.7 m, and only two main branches and one centre trunk of a first layer are reserved, therefore, the tree shape is called as the downward-arrowhead-shaped tree shape. The downward-arrowhead-shaped tree shape is small in top and large in bottom, the influence of upper branches on lower branches is less, and ventilation and light transmission conditions are good. The downward-arrowhead-shaped tree shape of a pear tree is an innovation based on the '3-plus-1' tree shape disclosed previously, which is simple in structure, convenient in culture, easy to grasp by fruit growers, convenient in field management, and easy to spread and popularize; and the tree shape is a novel high-quality and labour-saving cultivation tree shape of a pear tree.

The invention relates to palm print identifying technology, and provides a palm print identifying method and a palm print identifying device aiming at the defects that the preprocessing requirement for the acquired palm print images is high in the prior palm print identifying technology, and the like. The palm print identifying method comprises the step of extracting the palm print characteristics of an image to be extracted, and the extraction specifically comprises the following steps: rotating and correcting the image to be extracted to generate a corrected image; carrying out convolution operation for the corrected image according to the preset Gabor wavelet, carrying out the operation of taking an amplitude value or a phase from the corrected image after the convolution operation, and generating a response image; and carrying out local binary mode characteristic extraction for the response image to acquire a local binary mode code of the image to be extracted so as to generate the palm print characteristics. The invention also provides the palm print identifying device. The technical proposal of the invention can provide rotational invariance and has high processing speed, so the technical proposal can achieve real-time palm print identification.

The invention relates to the field of robot control methods, in particular to a multiple-gesture robot control method based on binocular vision. In order to overcome the defects that existing robot control methods based on vision are inconvenient to operate, hand identification is greatly affected by illumination and background colors, calculation workload of offline teaching methods is high, andthe precision requirements for robot model precision and coordinate determination are high, the multiple-gesture robot control method based on binocular vision is provided. The method comprises the steps that a binocular camera is arranged; a rectangular frame comprising gestures is manually selected; a classier is trained through a training sample set; a target is detected by the classier; then the target is tracked, and tracking results and detection results are fused; the deviation distance of the center point of the target from the initial point to the target point, a speed control instruction is output, and thus a robot translates; and characteristic points are extracted from the target frame, and rotary matrixes corresponding to the characteristic points are solved. The multiple-gesture robot control method is suitable for spraying robot control methods.

The invention discloses an image preprocessing method for processing nonuniform illumination of a miner face image and a coal surface. According to the image preprocessing method, by using a homomorphic filtering principle algorithm, illumination components are weakened and reflection components are enhanced, and meanwhile excessive inhibition to low frequency components is avoided. A homomorphic filtering function of the preprocessing method is used for separately processing the low frequency components and the high frequency components, and processing the low frequency components at different levels by segments, the inhibiting radius of the low frequency components is determined to be five pixel numbers according to the image filtering effect when the limiting frequency radius of the traditional low pass filter is five. The image preprocessing method has strong pertinence and remarkable effects to the processing of strong illumination influence in the miner face image and the coal surface.

The invention discloses an array substrate, a manufacturing method thereof and a display device, which relate to the technical field of display and aim at improving the use stability of the array substrate. The array substrate comprises a substrate and a thin-film transistor arranged on the substrate, wherein a light shielding layer is arranged between the thin-film transistor and the substrate; the light shielding layer comprises a light shielding metal layer and a light reflection adjustment layer formed on the substrate in a stacked mode; the light reflection adjustment layer covers the light shielding metal layer; and the reflectivity of the light reflection adjustment layer is lower than that of the light shielding metal layer. According to the array substrate, the manufacturing method thereof and the display device provided by the invention, the array substrate adopts the thin-film transistor in a top gate structure.

The invention discloses a brightness-adjustable car intelligent headlamp system based on the environment, and a method. The current environment of a car is preliminarily judged according to the car speed and light intensity, and the illumination brightness of the headlamp is adjusted based on the light intensity in the current environment, and the effects of energy saving and light pollution reduction are taken into account while meeting the lighting requirements. A design of the brightness-adjustable car intelligent headlamp system based on the environment is provided, and the brightness of the car lamp can be adjusted by collecting the environmental information through a sensor and judging by a master controller. The system comprises the sensor, the master controller and an actuator, thelighting brightness of the headlamp can be adjusted according to the driving state and the environmental information of the car, and energy is saved while providing a better road lighting effect fora driver.

The invention relates to a foam infrared image segmentation method based on NSST saliency detection and image segmentation, and the method comprises the steps: firstly carrying out the NSST decomposition of a foam infrared image, and obtaining a low-frequency sub-band image and a multi-scale high-frequency sub-band; secondly, performing saliency detection on the low-frequency sub-band image by adopting a GBVS algorithm to obtain a saliency value and a visual saliency map; thirdly, calculating a threshold value and a scale correlation coefficient for each high-frequency direction sub-band coefficient, and removing a noise coefficient, a non-linear enhanced edge coefficient and a weak edge coefficient; and finally, performing image segmentation on the NSST reconstructed image in combinationwith visual saliency to obtain a segmentation result. The method is strong in anti-interference capability and high in segmentation precision.

The invention discloses an elevator door repeated opening and closing detection method and system based on the computer visual technology. The elevator door repeated opening and closing detection method based on the computer visual technology comprises the following steps that a marker is pasted on the higher position of an elevator door; the marker is detected, and an elevator door image is collected to carry out door state analysis; according to the transmitted detection area, on an image gray channel, under the state that the door is completely closed, a background template is registered, the difference between a current frame column diagram at the current frame and a registered background frame column diagram at the registered background frame is subsequently and continuously calculated, and the opening and closing state of the door is judged; and through receiving each frame of door state type, and in combination with time domain statistics, whether the phenomenon of repeated dooropening and closing appears in a current elevator is analyzed, and a corresponding alarm signal is output.

The invention discloses a target detection tracking method based on a TLD. The method comprises the following steps of (1) at an initial frame of a video to be tracked, a tracking window assigned by a user forms a positive and negative sample pair detection module to carry out initialization training; (2) during a tracking process, a detection module and a tracking module work independently; the detection module calculates a LBP statistics characteristic vector of an image block accepted by a variance classifier and a merging classifier, carries out NCC nearest neighbor matching with a sample set and outputs a target position; and the tracking module uses a median optical flow method to predict a target position of a current frame through previous frame tracking; (3) an integration module integrates the detection module and the tracking module so as to carry out tracking result output, and for the currently-updated target position, a new positive and negative sample is generated so as to update the detection module; and (4) the step (2) and the step (3) are circulated till that tracking is ended. A comparison experiment performed on a public data set shows that by using the method, robustness and precision of the tracking can be increased to some extent.

The invention proposes an X-shaped corner extraction method based on an image contour, and relates to the technical field of visual detection. The method comprises the steps: sequentially traversing Harris corners on one contour through the construction of a window: taking a point which has the smallest coordinate mean value relative to seven corners as an X-shaped corner when the number of traversed corners is seven, carrying out the traversing of a next contour, extracting the X-shaped corner when the number of corners is four, and completing the extraction of X-shaped corners of all contours. The method can achieve the filtering of pseudo corners in an image, and can achieve the precise detection of a corner point in the image, wherein the detection precision can reach 0.09 pixels. Through the setting of the surrounding background of a corner, the method enables a mark point not to be affected by a feature point in an environment, and achieves the detection of the corner in a complex scene. Through combining with the advantages of Harris precise extraction and contour positioning, the method achieves the subpixel-level detection of a laid corner, and improves the coordinate precision of the corner. Through the dynamic adjustment of an image preprocessing threshold value, the method reduces the impact from the illumination in different scenes, and improves the corner detection success rate.

The invention belongs to the technical field of three-dimensional reconstruction of multi-sensor fusion, and provides a semantic live-action three-dimensional reconstruction method and system of a laser fusion multi-view camera. The method comprises the following steps: acquiring a multi-view camera image and correcting the multi-view camera image; acquiring laser point cloud data and visual inertial odometer data, and aligning the laser point cloud data with the visual inertial odometer data according to a timestamp to correct the laser point cloud data; interpolating the corrected laser point cloud data to obtain dense point cloud, projecting the dense point cloud to an imaging plane, matching the dense point cloud with pixels of the corrected multi-view camera image to obtain dense point cloud with RGB information of each frame, and superposing the dense point cloud to obtain real-view point clouds; and obtaining semantic information from the corrected multi-view camera image, matching the semantic information with the corrected point cloud to obtain instance object point cloud, and fusing the instance object point cloud with real scene point cloud to obtain a three-dimensional model for semantic real scene reconstruction.

The invention discloses an image classification algorithm and a system based on manifold learning. The method includes S1, selecting a training sample set and a testing sample set; S2, extracting feature points of image of two sample sets by using a sift algorithm; S3, respectively reducing the dimension of the feature points in the two sample sets by using the local linear embedding or Labras feature mapping in the manifold learning method; S4, inputting that reduced dimension feature point of the training sample set into a support vector machine classifier for training; S5, classifying the test sample set by using the trained support vector machine classifier. The invention combines the SIFT feature extraction algorithm with the non-linear manifold learning dimension reduction algorithm,extracts the middle-level feature of the image, and then classifies and processes the image by using the SVM classifier, thereby effectively improving the computing speed and the classification accuracy.

The invention discloses a tire pattern similarity detection method. According to the method, a tire pattern image is extracted based on template matching, and the pattern similarity is calculated by combining an improved LBP algorithm and a Hash algorithm. The algorithm comprises the following steps: extracting an overall tire image; carrying out tire pattern image extraction through optimized template matching; obtaining an LBP code with rotation invariance; extracting image fingerprint information, and calculating image similarity; segmenting the pattern image with a relatively high detection result, calculating local similarity, and screening out a part with high similarity. The method is high in adaptability, does not need to specify a tire image background, can accurately extract thetire pattern image, and solves the problems of variable tire pattern rotation angles, inconsistent illumination and the like.

The invention discloses a device and a method for collecting visible and near infrared spectrum of fruits in the field. A housing of the device is provide with a shutter and an integrating sphere, and the shutter is connected to a light source and is mounted above an entering hole of the integrating sphere, and the entering hole is located on an output probe, and the periphery of the output probe is mounted with an annular rubber gasket. A through hole below the integrating sphere is installed by a collimating lens, and the light of the collimating lens is transferred to a detector of a spectrum analyzer through an output optical to measure scene spectrum of the fruit in the field under the combined action of the spectrum analyzer light source and the scene ambient light and to measure field background spectrum of the fruit in the field only under the action of the scene ambient light. The scene spectrum and the field background spectrum are corrected to obtain visible and near infrared spectrum information which can really reflect the fruit quality. The device can effectively reduce the influence of field environment light, and is conducive to the application of visible and near infrared spectroscopy technology from the interior to the field, and can give full play to the visible and near infrared spectroscopy technology in the fruits of the field management and harvest maturity test.

The invention belongs to the technical field of chemical fiber industry, and particularly relates to a chemical fiber fluttering quality detection method and system based on a convolutional neural network. The system comprises an image acquisition and transmission module, a part detection positioning module, a parameter feedback module, a part recognition and judgment module, and a comprehensive judgment module. A CNN deep learning scheme is used for detecting the fiber fluttering abnormity, and a fiber fluttering abnormity detection technology of key part abnormity detection is provided. A dynamic feedback mechanism and a comprehensive evaluation mechanism are adopted; the mode of robot patrol camera snapshot and background analysis is used for replacing manual detection of abnormal conditions such as fiber fluttering and the like.

The invention provides a static gesture recognizing method and system for AR auxiliary maintenance. The method specifically comprises the following steps of: acquiring a video image and performing frame separation to obtain a test sample; separating a gesture from a background of an image in the test sample by using a skin color range method; creating a static gesture feature input vector corresponding to the image in the test sample and normalizing the vector; inputting the normalized the vector into a support vector machine and enabling a classifier in the support vector machine to recognize the gesture for acquiring an optimal classification solution of the static gesture. The method forms the eight-dimension composite characteristic input vector of the static gesture by the seven-order invariant moment of Hu and image normalization moment of inertia (NMI) so as to achieve better robustness for image translation, rotation, and zooming, fully takes account of the influence of illumination, background, and gesture orientation, increases the accuracy of training samples in the support vector machine, and increases recognition precision.

The invention discloses a levofloxacin lactate dispersible tablet. The dispersible tablet comprises, by weight, 100 parts of levofloxacin lactate, 40-45 parts of microcrystalline cellulose, 20-27 parts of sodium carboxymethyl starch, 38-42 parts of starch, 1-2 parts of micro-powder silica gel, 2-3 parts of magnesium stearate and 10-20 parts of ethanol water with the mass fraction of 60-70%, wherein the starch is hydroxypropyl starch; the micro-powder silica gel is silica gel modified by trimethyl chlorosilane. Part of auxiliary materials are optimized so that the usage quantity of the auxiliary materials can be greatly reduced, related substances, content, dissolution rate, dispersion uniformity and stability of the product are also improved, and therefore the aims of reducing the production cost and improving the product quality are achieved at the same time.

The invention provides a display device for simulation interaction based on an augmented reality technology, and relates to the technical field of virtual devices. Comprising a movable main body, a content generation module, a control terminal, an LED display screen arranged on the inner wall of the main body, a voice recognition module, a sound module, a camera used for capturing position information of a user and a somatosensory tracker used for dynamic capturing. The LED display screen is connected with the control terminal through the content generation module; the sound module, the voice recognition module, the camera and the somatosensory tracker are electrically connected with the control terminal. According to the technical scheme, a user can obtain corresponding content in a naked eye 3D interaction mode, the display limitation of physical space is broken through, and the equipment is portable, movable and reusable, so that the operation cost is low.

The invention relates to the technical field of pipeline detection, in particular to a pipeline foreign matter detection device and a detection method. The pipeline foreign matter detection device comprises a detection vehicle, a control part positioned in the detection vehicle, a mechanical arm and an image capture piece; one end of the mechanical arm is connected with the detection vehicle; and the image capture piece is connected with the other end of the mechanical arm. According to the pipeline foreign matter detection device and the detection method, metal foreign matters in a pipeline can be effectively and quickly detected, the influence of poor environment during pipeline detection and the inevitable problems of negligence, dangerousness, non-traceability and the like of manual detection are reduced, and convenience is provided for workers.

The invention provides a page content reading method and device based on image features. The method comprises the following steps: detecting whether the current page of the book is an updated page ornot; when the current page of the book is an updated page, triggering a device camera shooting end to shoot the current page of the book in a global shooting mode; otherwise, triggering a camera end of the equipment to shoot the current page of the book in a local shooting mode; obtaining page feature points corresponding to the current page of the book, searching page contents matched with the page feature points from a preset database; carrying out reading operation on the page content. Under the condition that the shooting frame rate of the shooting end is not changed, the pixel informationamount and the processing time needing to be processed when the device obtains the page feature points corresponding to the current page of the book according to the image frames are reduced; while the response speed of the equipment is increased, the power consumption of the equipment in the aspects of calculation and camera shooting is reduced, and the equipment requirement and the equipment burden are reduced.

The invention discloses a manufacturing method of a display panel. The manufacturing method comprises the following steps: providing a substrate; forming a first photoresist layer with pixel pits on the substrate through first exposure, wherein the material of the first photoresist layer comprises a negative photoresist material mixed with a photosensitive material; enabling the first photoresistlayer to be exposed for the second time to form a final photoresist layer, wherein the final photoresist layer comprises the remaining part of the first photoresist layer and a second photoresist layer located on the remaining part of the first photoresist layer, and the hydrophilic characteristics of the first photoresist layer and the second photoresist layer are different. The photosensitive material is mixed in the negative photoresist material, in the second exposure process, the upper layer part of the negative photoresist material receives more illumination, the photosensitive materialreacts under the illumination condition to become hydrophobic, and the lower layer part of the negative photoresist material is slightly influenced by illumination and still has the hydrophilic characteristic. When the color resistance material is injected into the pixel pits, ink can be effectively prevented from overflowing out of the pixel pits, and the probability of color mixing and mura phenomena is reduced.

The invention discloses an automatic guidance system and method based on radio frequency identification tags and vision. It adopts a four-wheel structure with two front-wheel steering and two rear-wheel drive. The DC motor is used as the walking drive device, and the steering is driven by the stepping motor, and a wheel angle positioning device is provided, which has a high dynamic response capability. Black and white two-color parallel guide belts are used as the guide path, and radio frequency identification tags are intermittently laid under it, with red station characters on both sides, CCD collects road surface information, and the brightness of the light source is adaptively adjusted according to the grayscale information of the image. Characters are extracted by difference, fast HOUGH transformation is used to identify straight lines, and the control signal is sent to the AGV controller after processing by the industrial computer to control the DC motor and stepper motor. In order to effectively realize the communication scheduling between AGVs in real time, radio frequency identification tags are used to realize positioning, and the wireless transceiver module realizes the communication between AGVs and the host. The invention has good environment adaptability and navigation accuracy, and the cost is low.

The invention discloses a high-altitude throwing detection method and device based on mixed vision, an event camera and a conventional camera are carried and registered at the same time to monitor a monitored building in real time, the event camera detects a moving target based on a generated event sequence, the position of the moving target is obtained, and the position of the moving target is detected; performing tracking correction on the position of the moving target by adopting a Kalman filtering algorithm, associating the detected moving target with a known moving target, establishing a tracking list of the moving target, judging whether the moving target is a high-altitude parabolic object, if so, reporting the position and time of the moving target, and according to the reported time of the moving target, determining whether the moving target is a high-altitude parabolic object. And intercepting and storing video images shot by the conventional camera. According to the invention, mixed vision of the event camera and the conventional camera is combined, the influence of illumination is small, and the false alarm rate is low and the positioning performance is high when a parabolic object is similar to a background in color, small in size, too fast in movement or at night.

The invention belongs to the field of decoration materials and in particular relates to environmental-friendly diatom ooze and a preparation method thereof. The diatom ooze consists of the following raw materials in parts by weight: 60-80 parts of modified refined kieselguhr, 10-20 parts of titanium dioxide gel, 6-8 parts of clay, 8-10 parts of attapulgite, 10-12 parts of kaolin and 20-30 parts ofwater. The preparation method of the environmental-friendly diatom ooze comprises the following steps: (1) loading nano titanium dioxide by gaps and surfaces of the modified refined kieselguhr; (2) uniformly mixing the modified refined kieselguhr, the clay, the attapulgite, the kaolin and the water; (3) loading the cerium oxide and the indium oxide by the gaps and the surfaces of the mixture. Theenvironmental-friendly diatom ooze provided by the invention has high light reflection efficiency because of micro pores inside, the lighting intensity in the micro pores is improved, the optical energy utilization rate of the nano titanium dioxide can be increased, a good catalysis effect can be achieved, a rapid formaldehyde decomposition speed is achieved, and the formaldehyde decomposition effect is slightly affected by lighting.

The invention discloses a monocular camera object identification and positioning method based on a vector illumination influence removal algorithm, and the method comprises the steps of S1, collecting an image map through a robot, and carrying out illumination removal processing on the image map; S2, filtering the image map after illumination removal processing, and obtaining a filtered image; S3, performing color space conversion on the image, and obtaining a binary image according to the color threshold value of the target object; S4, eliminating interferents outside target area, determining a target object, and obtaining the image coordinates of the target object; S5, judging whether the target object exists in the target area or not according to the image coordinates of the target object, and if not, returning to null and ending the identification and positioning; if yes, returning to the center coordinate of the target object, then adjusting the head angle and the body posture of a robot to enable the center coordinate of the target object to coincide with the center position of the image, then solving the distance from the target object to the robot according to a calculation model, and obtaining the recognition result of the target object.

The invention provides a marker detection method for the image recognition of target points in natural environments. According to the invention, for realizing the rapid and automatic image detection of target points in a natural environment, blue-yellow-alternated markers are designed, and through methods such as line scanning, color-difference enhancement, contour tracking and geometric parameter measurement and the like, the accurate recognition and positioning on the markers are realized. Experimental results show that the method disclosed by the invention can accurately and rapidly complete the recognition and positioning of a to-be-measured target point in outdoor natural environments including various conditions such as moderate illumination, dark illumination, perpendicular incidence of highlight, shadow interference and night and the like. The detection method has the advantages that the method is accurate, rapid, stable, less in illumination influence, and strong in adaptability, and the like.

The invention discloses an elevator door repeated opening and closing detection method and system based on the computer visual technology. The elevator door repeated opening and closing detection method based on the computer visual technology comprises the following steps that a marker is pasted on the higher position of an elevator door; the marker is detected, and an elevator door image is collected to carry out door state analysis; according to the transmitted detection area, on an image gray channel, under the state that the door is completely closed, a background template is registered, the difference between a current frame column diagram at the current frame and a registered background frame column diagram at the registered background frame is subsequently and continuously calculated, and the opening and closing state of the door is judged; and through receiving each frame of door state type, and in combination with time domain statistics, whether the phenomenon of repeated dooropening and closing appears in a current elevator is analyzed, and a corresponding alarm signal is output.