31results about How to "Automatic and accurate detection" patented technology

The disclosure discloses a printer and a method and device for controlling the same. The method for controlling the printer comprises: detecting the number of heating units of a currently mounted printing head; acquiring a resolution, corresponding to the detected number of the heating units, of the printing head; and setting a printer parameter according to the acquired resolution of the printing head. The problem of high error rate after the resolution of the printing head of the printer is changed is solved.

The invention relates to a testing device for an undervoltage tripper, which comprises a programmable power supply with maximal output voltage being no less than square root of 2 times of rated working voltage U of the undervoltage tripper, a force measuring device and a computer; the computer is connected with the programmable power supply, and used for sending out a control signal of the programmable power supply and receiving a voltage signal of the programmable power supply; the programmable power supply is connected with a coil for an undervoltage tripper to be tested; the detection signal output terminal of the force measuring device used for detecting the tripping force of the undervoltage tripper to be tested is connected with the computer; a tripping force measuring part of the force measuring device is arranged oppositely to an armature for the undervoltage tripper to be tested. The force of releasing tripping can be detected during the process of detecting the power-on or power-off response time and the pull-in or release drop-away voltage of the undervoltage tripper, and the testing device can automatically finish measuring all parameters of the undervoltage tripper without manual intervention, therefore, the large-scale mass detection can be carried out.

Embodiments of the present invention provide a tire pressure detection method and a computer storage medium. Wherein, the method includes: performing a feature extraction operation on the wheel and tire image to be detected to obtain tire feature data of the wheel and tire image; based on the first feature data included in the tire feature data, performing Preliminary judgment, obtain the preliminary judgment result of the tire pressure of the wheel tire; if the tire pressure of the wheel tire is determined to be normal based on the preliminary judgment result, then based on the second feature data included in the tire feature data, the tire pressure of the wheel tire in the wheel tire image Prediction is performed to obtain a prediction result of the tire pressure of the wheel and tire; based on the prediction result of the tire pressure of the wheel and tire, the tire pressure of the wheel and tire in the image of the wheel and tire is re-judged to obtain a re-judgment result of the tire pressure of the wheel and tire. This embodiment can automatically detect the tire pressure, thereby reducing the possibility of accidents when the vehicle is running at low speed and high speed.

The invention relates to a system for automatically testing parameters of an under-voltage trip unit. The system comprises a programmable power supply, a force detection device and a computer, wherein the computer is connected with the programmable power supply and is used for transmitting a control signal of the programmable power supply and receiving a voltage signal of the programmable power supply; the programmable power supply is connected with a coil of an under-voltage trip unit to be detected; a detection signal output end of the force detection device which is used for detecting trip force of the under-voltage trip unit to be detected is connected with the computer; a trip force testing part of the force detection device is arranged opposite to an armature of the trip unit to be detected; and a detection device for detecting the suction state or the release state of the armature of the under-voltage trip unit to be detected is arranged at the armature of the under-voltage trip unit to be detected. By the system, the released trip force can be detected while the power acquisition or power loss response time and the suction or release trip voltage of the under-voltage trip unit are detected; and by the device, all the parameters of the under-voltage trip unit can be automatically measured without manual interference, so that the testing parameters can be detected on a large scale in batches.

An image processing method to sample the image to generate patches. Feature vectors are extracted from the patches, and the extracted feature vectors are partitioned into clusters, where feature vectors in the same cluster share a common characteristic. A portion of interest in the image is segmented. An aggregate bounding region creation process is carried out by finding the largest segment and creating a bounding box around it; determining which cluster contains the most patches within the bounding box of the segment; and adding the patches of the determined cluster to an aggregate bounding region for the portion of interest. The aggregate bounding region creation process is repeated for each other segment in order of size. The resulting aggregate bounding region contains all the patches associated with the portion of interest. The patches which fall outside the resulting aggregate bounding region are then removed from the image.

The invention discloses a railway engineering detection method and system based on an unmanned aerial vehicle inspection video. The method includes: detecting the inspection video frame by frame based on a target detection algorithm and obtaining a plurality of target detection frames including target structures; The target tracking algorithm tracks each target detection frame and obtains the target trajectory of each target structure; determines the estimated coordinates of each target structure according to the frame data of each trajectory frame in each target trajectory; The design data of each target structure in the inspection video determines the positioning information of each target structure in the inspection video; among them, the inspection video is based on the video of the whole or part of the predetermined railway line shot by the drone, and the frame data includes unmanned The position data of the drone and the attitude data of the camera can be used to automatically and accurately detect the target structures on the railway line from the drone inspection video, thereby improving the efficiency of railway construction inspections.

The invention discloses a method for adjusting the direction of equipment according to the wind direction, which includes: detecting the wind direction through an accelerometer; and adjusting the direction of the equipment according to the detected wind direction. Correspondingly, the invention also discloses a device for adjusting the direction of the equipment according to the wind direction. The technical solution adopted by the invention can automatically detect the wind direction in real time and dynamically adjust the direction of the equipment.

The present invention relates to the technical field of optical fiber signal detection, in particular to a high-magnetic field transmission performance testing device for optical fibers, which includes: a frame; a stretching drive mechanism, which is horizontally arranged on the frame; Two sets of optical fiber clamping mechanisms are respectively set on the two output ends of the stretching drive mechanism; there are two sets of end push-out mechanisms, and two sets of optical fiber clamping mechanisms are respectively set on the two output ends of the stretching drive mechanism On the top, the terminal push-out mechanism is located between the output end of the stretching drive mechanism and the optical fiber clamping mechanism; the electrical signal connection mechanism; the magnetic field generating device; On both sides of the stretching drive mechanism, a strong magnetic field transmission performance testing device for optical fibers shown in the present invention can automatically stretch the installed optical fiber, and can comprehensively reflect the difference between the magnetic field and optical fiber signal transmission performance under different conditions. relationship between.

The invention discloses a method for automatically detecting facial flaws. The method is applied to intelligent mirror equipment with an image capturing module, a mirror surface and a processing unit.The method comprises the steps of capturing a face image of a user through the image capturing module when the user is in front of the mirror surface, generating a smooth image through the processingunit according to the face image, generating an epidermis change image according to the difference between the two images, identifying flaws in the epidermis change image and recording the positionsof the flaws. According to the method, the position of the facial flaws can be automatically and accurately detected.

The invention discloses a device and method for searching for a loop channeling electric fault in direct current systems. The device comprises an insulation monitoring device, wherein the insulation monitoring device comprises a processer, a bus connection interface, an axle type resistance control circuit, an adjustable resistance control circuit and a current transformer. The method includes the steps that according to the difference between the voltage to ground of the positive electrode of the first direct current system and the voltage to ground of the positive electrode of the second direct current system, and therefore whether the loop channeling electric fault exists between the two direct current systems or not is judged; when the loop channeling electric fault exists between the two direct current systems, an adjustable resistor is connected in according to the voltages to ground of the positive electrodes and the negative electrodes of the direct current systems, and the resistance value of the adjustable resistor is adjusted; current data acquisition is conducted on the current transformer connected to a feeder wire, and the feeder wire where the loop channeling electric fault occurs is positioned according to the preset change rules and the change rules of the obtained current data. By means of the device and method, the loop channeling electric fault can be found and positioned rapidly, accurately and safely. The device and method can be widely applied to the field of electric systems.

An on-line detection method for the optical properties of a photochromic film, comprising using a conveyor belt to transport a polyester film, irradiating and collecting a discolored image on the upper side of the conveyor belt, and fading and collecting the faded image on the lower side of the conveyor belt, and sending the image to an image recognition model In this way, the areas where the optical properties of the polyester film do not meet the requirements are marked on the line. The method has the advantages of convenient operation, high precision and fast detection speed, and can meet the strict requirements of on-line detection of production lines.

An image processing chip test method comprising: controlling a power supply circuit to provide a first operating voltage to an image processing chip comprising a storage device; and when reading written first image data from the storage device, the test device receives a first error detection code corresponding to the first image data and determines whether the first error detection code means an error occurs. If an error occurs, record the first operating voltage as an erroneous operating voltage, and if the error does not occur, provide a second operating voltage to the image processing chip. Also, when the written second image data is read from the storage device, the test device receives a second error detection code corresponding to the second image data and determines whether the second error detection code means an error occurs.

The invention discloses a plant nematode detection method and system based on deep learning and multi-feature combination. The present invention uses the first deep learning neural network to extract the key features of plant nematodes, and respectively detects the key characteristic regions such as the head, female tail, and male tail of the plant nematodes; and uses the second deep learning neural network as the identification and detection network to achieve Single-feature detection, dual-feature joint detection and three-feature joint detection have high detection accuracy. Finally, a humanized nematode detection interface is designed, and an intelligent nematode detection system is completed. The nematode detection and identification scheme can be selected according to the degree of difficulty in distinguishing nematodes. The invention adds a mechanism for effectively extracting key features, and a method for multi-feature joint classification. The invention can accurately detect plant nematode data with small key feature points and are difficult to distinguish, and utilizes multi-feature combination to realize hierarchical detection of pine wood nematodes and pine wood nematodes.