Patents
Literature

1095results about How to "Improve adaptability" patented technology

Upper limb rehabilitation training robot

The invention discloses an upper limb rehabilitation training robot. The upper limb rehabilitation training robot comprises a pedestal, an ergonomic seat, a rehabilitation mechanical arm and a man-machine interaction control cabinet, wherein the pedestal is provided with a height adjusting mechanism and a transversing mechanism; the transversing mechanism is provided with a suspending arm; the rehabilitation mechanical arm comprises a shoulder abduction/adduction joint, a shoulder bending/stretching joint, an upper arm, an elbow joint, a front arm inward turning/outward turning joint, a front arm, a wrist bending/stretching joint, a palm and fingers, wherein the shoulder abduction/adduction joint is arranged on the suspending arm; the shoulder bending/stretching joint is connected with the shoulder abduction/adduction joint; the front arm is connected with the upper arm by the elbow joint and provided with the front arm inward turning/outward turning joint; the palm is connected with the front arm by the wrist bending/stretching joint and connected with the fingers by a connecting rod; the upper arm, the front arm and the palm are fixedly connected with three force sensors; and the sensor fixedly connected to the upper arm is a two-dimensional sensor. The robot provides composite rehabilitation motion for each joint of the upper limb of a patient in a three-dimensional space and can finish the rehabilitation training of the upper limb with high quality.
Owner:付风生 +1

Methods, systems, and computer program products for compressing a computer program based on a compression criterion and executing the compressed program

Embodiments of systems, methods, and computer program products are provided for compressing a computer program based on a compression criterion and executing the compressed program. For example, a computer program may be compressed by scanning an initial computer program to identify one or more uncompressed instructions that have a high frequency of use. A storage mechanism, such as a data structure, may then be populated with the identified uncompressed instructions. A compressed computer program may be generated by respectively replacing one or more of the identified uncompressed instructions with a compressed instruction that identifies a location of the corresponding uncompressed instruction in the storage mechanism. Additional compression of the computer program may be achieved by scanning the compressed computer program to identify one or more uncompressed instructions that have a high frequency of use when at least a portion of their instruction operand is ignored. A second storage mechanism, such as a data structure, may then be populated with the identified uncompressed instructions. Finally, a further compressed computer program may be generated by respectively replacing one or more of the identified uncompressed instructions with a second type of compressed instruction that identifies a location of the corresponding uncompressed instruction in the second storage mechanism.
Owner:MACOM TECH SOLUTIONS HLDG INC

Semi-supervised-transfer-learning character recognition method and system based on convolutional neural network

The invention provides a semi-supervised-transfer-learning character recognition method based on a convolutional neural network. The method includes: using batch character image samples, which are ina target domain and have no class label, as a test sample set to input the same to the convolutional neural network after semi-supervised transfer learning, and identifying character images of the test sample set, wherein the convolutional neural network after semi-supervised transfer learning is obtained by joint training of batch character image samples which are in a source domain and have class labels, batch character image samples which are in the target domain and have class labels, and batch character image samples which are in the target domain and have no class label, and recognitionprecision is improved. According to the semi-supervised-transfer-learning character recognition method based on the convolutional neural network provided and a semi-supervised-transfer-learning character recognition system based on the convolutional neural network provided by the invention, the large number of source-domain samples with the class labels, the small number of target-domain sampleswith the class labels and the relatively easily obtained target-domain samples without a class label can be utilized for semi-supervised transfer learning, and adaptation ability of the convolutionalneural network on the target-domain samples is improved.
Owner:TSINGHUA UNIV

Feature image extraction method based on deformable convolutional layer and feature image extraction device thereof

The invention discloses a feature image extraction method based on a deformable convolutional layer. The method comprises the steps that a target image is acquired; and pixel values are extracted from the target image through the sampling points of the convolutional kernel in the convolutional layer so as to obtain a feature image, wherein the actual coordinate values of the sampling points are the actual coordinate values calculated according to the preset initial coordinate values and the deviation variable trained in advance. According to the method, one deviation variable can be additionally arranged for each sampling point in the convolutional kernel so that a convolutional neural network is enabled to realize the capacity of learning image space geometric deformation, the convolutional kernel is enabled to randomly sample near the present position, the adaptability of the convolutional layer for the deformed image in extracting the feature image can be increased and the identification accuracy for the deformed object can be increased. The invention also discloses a feature image extraction device based on a deformable convolutional neural network. The feature image extraction device also has the beneficial effects.
Owner:GUANGDONG UNIV OF TECH

Inertial vision integrated navigation method based on optical flow method

The invention discloses an inertial vision integrated navigation method based on an optical flow method, relates to an inertial vision integrated navigation method, and has the purpose of solving theproblems that the navigation information is inaccurate due to cumulative errors caused by the existing inertial navigation working for a long time, a mask interruption or an excessive dynamic error may happen when the existing satellite navigation is indoor or shielded by a building, and the algorithm of the existing vision navigation is complex and susceptible to camera pose, illumination changes, image noise and the like. The method comprises the following processes: 1, defining a coordinate system; 2, carrying three sensors of an IMU (Inertial Measurement Unit), a camera and an altimeter onthe drone to obtain two-dimensional optical flow data; 3, obtaining the position of the drone in the world coordinate system; 4, performing inertial navigation according to the IMU measuring information to calculate the position and attitude of the drone in the world coordinate system; and 5, obtaining the position and attitude information of the merged drone in the world coordinate system. The method is used in the technical field of autonomous navigation of drones.
Owner:HARBIN INST OF TECH

Laser flexible micro-processing system and method based on liquid crystal spatial light modulator

The invention discloses a laser flexible micro-processing system and method based on a liquid crystal spatial light modulator. The system comprises a laser, a first reflecting mirror, a half-wave plate, a beam expander, a second reflecting mirror, a first spatial light modulator, a second spatial light modulator, a quarter-wave plate, a first lens, a third reflecting mirror, a spatial filter, a fourth reflecting mirror, a second lens, a beam splitter, a third lens, a CCD camera, a galvanometer, a field lens, a working table and a computer. With the utilization of the wavefront modulation characteristic of the liquid crystal spatial light modulator, a widely adaptable flexible laser processing technology which is controllable in multiple parameters such as laser beam quantity, beam shape and quality, polarization state and angular momentum is provided; in conjunction with wavefront measurement and feedback technology, not only can wave front aberration be corrected, but also a high-energy beam can be decomposed into random two-dimensional or three-dimensional low-energy multiple beams, and parameters of the beams such as shape and polarization state can be flexibly controlled, so that the laser processing technology has the advantages of high quality, high efficiency, extremely high adaptability and flexibility.
Owner:HUBEI UNIV OF TECH

Image acquisition and processing method and device of paper money discriminator

The invention discloses an image acquisition and processing method of a paper money discriminator. The image acquisition and processing method comprises the following steps of: (a) acquiring a paper money image line by line by using a linear array image sensor; (b) performing analog-digital conversion on the acquired paper money image by using an image digitalization module and correcting a digital signal after conversion; (c) outputting the corrected digital signal to a digital signal processor to detect the paper money image; and (d) controlling stop and start of a paper money transmission motor when non-abnormal paper money is detected, wherein a process of acquiring the paper money image is controlled by using a coded disc signal in the step (a). The invention also discloses an image acquisition and processing device of the paper money discriminator, wherein the image acquisition and processing device comprises the linear array image sensor, the image digitalization module, the digital signal processor, the paper money transmission motor and the image acquisition control module; the linear array image sensor is used for acquiring the paper money image line by line; the image digitalization module is used for performing analog-digital conversion and signal correction; the digital signal processor is used for performing paper money image detection, the paper money transmission motor is used for controlling the paper money transmission; and the image acquisition control module is used for controlling the process of acquiring the paper money image by using the coded disc signal.
Owner:ZHEJIANG FOTA ELECTRIC

Compound vibration drilling device and machining method

The invention discloses a compound vibration drilling device and a machining method. The compound vibration drilling device comprises machine tool spindle motor and a drill bit and is characterized by further comprising a vibration tool handle and a vibration control system. The vibration control system comprises a monitoring system and an executing system. The executing system comprises a low-frequency vibration generator, an ultrasonic vibration generator, a stepping motor, a servo drive motor and an electric brush. The vibration controller controls the low-frequency vibration generator through the servo drive motor and the stepping motor. The vibration controller controls the ultrasonic vibration generator through the electric brush. By means of the machining method, on the basis of traditional laminated component drilling machining, the compound ultrasound or low-frequency axial vibration is adopted according to material attribute differences of different materials, conversion of vibration drilling and common drilling, conversion of ultrasonic vibration drilling and low-frequency vibration drilling and adjusting of vibration parameters in the machining process can be achieved through the vibration controller, and high-quality manufactured holes are obtained.
Owner:NANJING UNIV OF AERONAUTICS & ASTRONAUTICS

Color correction method for RGB-IR (Infrared) image sensor based on infrared environment

The invention provides a color correction method for an RGB-IR image sensor based on an infrared environment. The method comprises the following steps: step 1), inputting an RGB-IR image from the RGB-IR image sensor, performing interpolation processing on the RGB-IR image, calculating and obtaining numerical values of RGB channels and IR channels at each pixel location, and obtaining full format RGB and IR images; step 2), subtracting numerical values of the IR channels at corresponding positions from the numerical values of R, G, and B channels, obtaining a numerical value of a practical RGB channel at each pixel location, and then obtaining an RGB channel value whose infrared components are filtered; and step 3), performing WB (White Balance) color temperature and CCM (Color Correction Matrix) color correction on the R, G and B channels of step 2), and outputting a corrected RGB image. The color correction method for the RGB-IR image sensor based on the infrared environment provided by the invention has the advantages that the influence of multiple infrared components on the RGB channels can be eliminated in real time without using an IR-CUT, and the influence of the color temperature and color of visible lights on the RGB channels can be corrected; the infrared components can be separated in real time through a direct calculation method, and a correction matrix is not needed to be calculated complicatedly, thus the steps are more concise, the controllability is higher, and the adaptability of the infrared environment is better.
Owner:杭州雄迈集成电路技术股份有限公司

Range hood capable of self-adaptively regulating rotating speed of fan

The invention discloses a range hood capable of self-adaptively regulating the rotating speed of a fan. The range hood comprises a range hood shell, a motor arranged in the range hood shell and a speed regulation control circuit capable of controlling the rotating speed of the fan, wherein the motor is connected with an air outlet through an exhaust pipeline; and the air outlet is communicated with a public flue. The range hood is characterized by being provided with a sensor capable of detecting the pressure variation of the public flue, wherein an induction output end of the sensor is connected with an input end of the speed regulation control circuit; and an output end of the speed regulation control circuit is connected with the motor. When the pressure variation of the public flue is detected by the sensor, a corresponding decision is made; according to output power of a motor system which is required under corresponding pressure, the rotating speed of the fan of the range hood can be self-adaptively regulated, so that the smoke exhausting effect of the public flue is improved, and the flue is prevented from being plugged; and simultaneously, cooking fumes in a room can be smoothly removed, and a smoke exhaust environment of a kitchen is effectively improved.
Owner:NINGBO FOTILE KITCHEN WARE CO LTD

Automatic tile sticking device for building construction

The invention discloses an automatic tile sticking device for building construction. The automatic tile sticking device comprises a bending plate connected with a roller, one end of the bending plateis fixedly connected with a U-shaped frame, the U-shaped frame is internally and rotatably connected with a threaded rod, the U-shaped frame is connected with a power mechanism, a siding block sleevesthe threaded rod in a threaded mode, the sliding block is connected with the U-shaped frame through a limiting mechanism, a platform plate is arranged at one end of the sliding block, the platform plate is fixedly connected with the sliding block through a cross rod, the two ends of the platform plate are fixedly connected with connecting plates, and the opposite sides of the two connecting plates are fixedly connected with fixed blocks, and sliding rods are fixedly connected between the two fixed blocks and the platform plate. According to the automatic tile sticking device for the buildingconstruction, the height of tile sticking can be adjusted according to needs, adaptability of the tile sticking device is greatly improved, the intensity of labor is reduced, the occurrence of dangersis reduced; and tiles needing to be stuck are fixed and stuck, and the sticking efficiency is improved.
Owner:湖南望轶工程设备租赁有限公司

Robot-based gluing quality on-line visual detection device and method

ActiveCN106066328AExtend working hoursImprove adaptabilityMaterial analysis by optical meansLight sourceVisual detection
The invention belongs to the field of automobile gluing and detection and discloses a robot-based gluing quality on-line visual detection device and method. The device comprises a camera, a light source, a computer, a glue gun, a robot and a light source controller. The method comprises dividing a gluing locus into several sections through robot program, outputting glue section numbers to the computer through the robot program, selecting camera and light source parameters, light source controller and computing coefficients and a glue bond diameter limit through the computer according to the glue section numbers, taking a picture through the camera, storing the image, and calculating a glue band diameter according to a glue band diameter-direction pixel pitch extracted from the image and the computing coefficients, storing the glue band diameter through the computer, comparing the stored glue band diameter and the selected glue bond diameter limit and displaying an alarm if the stored glue band diameter is greater than the glue bond diameter limit. The device has a simple structure. The device and method are economic and practical, realize accurate detection and solve the problem that the existing detection device cannot control multiple visual detection units.
Owner:CHINA FIRST AUTOMOBILE
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products