16results about How to "Reduce detection error rate" patented technology

The invention provides an image feature detection method based on flexible connection, relates to the technical field of image analysis and processing, and aims to solve the defects of an existing image feature detection method in the aspects of accuracy, processing speed, detection sensitivity, false detection rate and the like. The method comprises the following steps: firstly, acquiring an input image, converting the input image into a grayscale image, carrying out Gaussian filtering processing, then calculating a gradient magnitude and direction, and carrying out non-maximum suppression and dual-threshold processing to obtain a preliminary edge image and a final edge image; and then connecting the weak edge pixels, extracting image features through a connected region analysis algorithm, calculating parameters such as a geometric center, a perimeter and an area, judging whether the features meet requirements or not according to a preset rule, and finally outputting the image features meeting the requirements. According to the method, parameters can be flexibly adjusted, different images can be adapted, the method is excellent in the fields of object recognition, medical image analysis, industrial detection and the like, the detection accuracy and efficiency are improved, the false detection rate is reduced, and the method has wide application prospects and practical value.

The present invention relates to the field of Internet information processing technology, and specifically to a sensitive information detection method based on a deep model and a knowledge base, S1. Preprocessing text information input by a user; S2. Detecting sensitive information on the text information through a BERT deep model and a knowledge base; S3. Filtering the sensitive information detection results; S4. Merging and outputting the filtered sensitive information. The present invention can improve the accuracy of sensitive information detection.

The invention relates to a signal processing method, in particular to a signal saliency analysis method based on a local noise amplitude probability model, and aims to solve the problems that when signal detection is carried out by using the prior art, the background noise amplitude of a signal is relatively large, so that a detection condition is mistakenly triggered, and a detection result is influenced. In the operation process, firstly, an original time sequence waveform containing a signal to be detected is determined and subjected to primary processing, the linear trend of the original time sequence waveform is removed, and filtering is carried out; secondly, respectively selecting a noise amplitude statistical window and one or more signal windows comprising to-be-detected signals in the filtering waveform, and respectively calculating the root-mean-square amplitude of noise and the root-mean-square amplitude of the signals in the signal windows; then, calculating the probability that the root-mean-square amplitude of the noise reaches or exceeds the root-mean-square amplitude of the signal in the signal window; and finally, setting a judgment condition, and judging whether the to-be-detected signal is significant relative to background noise or not according to the obtained probability.

The utility model discloses a microminiature heterosexual battery detection robot, which comprises a bottom base, a robot main body and a detection mechanism, the robot main body is arranged on the upper surface of the bottom base, the detection mechanism is arranged on one side of the robot main body, the upper end of the robot main body is movably connected with a rotating joint, and the rotating joint is movably connected with the bottom base. And a mechanical arm is arranged on the other side of the rotating joint, an execution clamp is arranged at the other end of the mechanical arm, a first visual sensor is arranged at the upper end of the interior of the detection mechanism, and a protective cover is connected to the outer side of the first visual sensor in an embedded mode. Through the arrangement of the detection mechanism and the robot main body, the device does not need to manually detect the surface structure of the battery, the battery detection efficiency is improved, and the wind power output by the fan in the detection mechanism can clean dust and impurities on the surface of the microminiature heterosexual battery. And the probability of detection errors caused by dust and impurities in the detection process of the microminiature heterosexual battery is reduced.

The present application discloses a quality inspection control method for biological in vitro samples and a computer storage medium. The method mainly transfers the carrier of the biological in vitro sample between the transportation area and multiple work areas of the quality inspection workbench according to the detection process through a transfer step to perform a first diversion treatment on it; detects the sample through the quality inspection step to generate quality inspection data, and transmits the quality inspection data to the quality inspection management system, and performs a second diversion treatment on the carrier; transports the carrier of the completed quality inspection sample out of the quality inspection workbench or transports the empty carrier located in the screening area out of the quality inspection workbench for standby use; the method of the present application determines the detection process of the biological in vitro sample and the determination of the verification results during the detection through information interaction with the quality inspection management system, which can effectively improve the detection accuracy and applicable scenarios, and the devices in the quality inspection workbench work together to form a standardized processing process for the sample, thereby improving the processing efficiency.

The application discloses a method for determining the oil adsorption capacity of shale, comprising the following steps: (1) pretreatment of a shale sample; (2) preparation of a crude oil solution; (3) placing a pretreated shale powder sample with a mass of m1 in the crude oil solution for an adsorption experiment; (4) centrifugal weighing to obtain the mass m2 of the shale powder sample in a liner tube; and (5) calculating the shale adsorption capacity Sp by using a model for determining the oil adsorption capacity of shale, wherein the model is shown in formula (1): Sp=(m2-m1) / m1 (1), wherein Sp represents the model for determining the oil adsorption capacity of shale, m2 represents the mass of the shale after oil adsorption, and m1 represents the mass of the shale before oil adsorption. The application further discloses an application of the method for determining the oil adsorption capacity of shale in shale oil exploration.

The invention provides an automatic detection method and device for a photovoltaic inverter, electronic equipment and a storage medium, and the method comprises the steps: responding to an input parameter and a judgment threshold parameter which are configured for a to-be-detected photovoltaic inverter in a detection configuration interface by a detection user under a target detection unit; detecting an actual input parameter and an actual output parameter of the to-be-detected photovoltaic inverter for the input parameter under the target detection unit through external detection equipment communicating with the to-be-detected photovoltaic inverter; and performing numerical comparison on the difference value between the actual input parameter and the actual output parameter and the judgment threshold parameter to obtain a detection result of the to-be-detected photovoltaic inverter for the input parameter under the target detection unit. According to the invention, the test efficiency and the production efficiency can be improved, and the detection error rate is reduced.

The invention discloses a detection image viewing method and system for chip bonding line detection. The detection image viewing method comprises the following steps: acquiring a detection image set consisting of a plurality of detection images shot by a plurality of different shooting distances at the same detection position of a detected chip bonding line; selecting a corresponding image transformation area on a first detection picture displayed on the current image interface, switching a first image in the current image transformation area to images in other detection pictures according to the image switching instruction, and keeping the image, displayed outside the image transformation area, of the current interface unchanged, by replacing the frame selection area in the currently displayed picture with the corresponding area image in other pictures with different focal lengths, a detector can identify and analyze the connection state of the detected metal wire by observing and comparing the switched image in the frame selection area and the current picture outside the frame selection area, so that the detection efficiency is improved.

This application provides an obstacle detection method, apparatus, electronic device, and machine-readable storage medium. The method includes: acquiring binocular depth images using a binocular camera deployed on a mobile robot; extracting obstacle contours based on the binocular depth images to obtain initial obstacle contours; for any obstacle contour, determining a first obstacle ROI region in a first grayscale image and a second obstacle ROI region in a second grayscale image corresponding to the obstacle contour; determining a first disparity value of the obstacle corresponding to the obstacle contour based on the first and second obstacle ROI regions; and determining whether the obstacle corresponding to the obstacle contour is a credible obstacle based on the first and second disparity values. This method can improve the reliability of obstacle detection and reduce the implementation cost of obstacle detection.

The utility model provides a color broadcast detection device. The color broadcast detection device comprises a bearing plate and a control circuit arranged on the bearing plate, the control circuit comprises a sound controller, a plurality of switches and a connecting seat; the sound controller is provided with a VCC pin, a GND1 pin, a plurality of OUT pins and a GNDSW pin; the connecting seat is provided with a power supply pin and a plurality of signal pins; the VCC pin is connected with a positive electrode of a power supply, and the GND1 pin is connected with a negative electrode of the power supply; the power pin is connected with a power anode; one end of each switch is connected with one OUT pin and one signal pin, the other end of each switch is connected with the power supply cathode, and the GNDSW pin is connected with the other end of each switch and is connected with the power supply cathode. Compared with the prior art, the light color detection device has the advantages that the sound controller is matched with the switch to detect the light color, the technical problem that defective products are not detected due to the fact that detection personnel easily remember the light color which should be lighted by the light mistakenly when multi-color light detection is carried out manually in the prior art is solved, and the probability that the detection personnel carry out mistaken detection is effectively reduced.

The present invention discloses an organosilicon surfactant cloud point measurement apparatus, and relates to the technical field of chemical sample analysis, the organosilicon surfactant cloud point measurement apparatus comprises a pollution discharge assembly, a heating stirring assembly and a detection assembly, the heating stirring assembly is arranged on the outer side of the top of the pollution discharge assembly, and the detection assembly is arranged on the outer side of the top of the heating stirring assembly. Clear water contained in the transparent test tube does not make contact with a solution in the detection bottle, so that the clear water can be kept clear all the time, the inner contour and the outer contour of the piston block are attached to the transparent test tube and the detection bottle respectively, and the light guide layer is arranged at the middle end in the piston block. Through the design, when the solution in the detection bottle reaches the cloud point and changes color, detection personnel or visual detection equipment can confirm that the organosilicon surfactant reaches the cloud point through the obvious color difference between the light guide layer end and the solution side, so that the detection accuracy is improved, and the detection efficiency is improved. Therefore, the cloud point measurement is efficient and simple while high precision is ensured.

The invention discloses a plastic optical lens surface defect detection method and system, and the method comprises the steps: S1, carrying out the phase consistency multi-scale edge detection of a plastic lens image, and obtaining a coordinate set of edge points of a plastic lens; s2, fitting an elliptic model based on the edge point set, dividing an optical region and a non-optical region by using a fitting result, and generating an optical region mask; s3, carrying out enhancement and feature extraction on pixels in the optical region, screening a potential defect region through binaryzation to obtain a potential defect map, and obtaining the length and width of the defect region by using a quadratic element measuring instrument; s4, constructing a defect detection model based on VGG16, and training and optimizing the defect detection model by using the labeled training sample to obtain an optimized model; and S5, detecting the plastic lens image of which the defect category is not marked by using the optimized model to obtain a defect detection result. According to the invention, false detection and missing detection of surface defects of the plastic optical lens can be reduced.

This invention discloses a camera detection method and system based on electromagnetic radiation signals. The method includes the following steps: acquiring radio frequency signals from the environment and performing preprocessing and spectrum analysis; detecting and marking peak points in the spectrum; grouping the marked peak frequencies and calculating the basic interval frequency of each group; determining whether the basic interval frequency of each group matches the camera clock frequency range; if a match is found, retaining the corresponding peak frequency group; discarding or merging peak frequencies; and further determining whether horizontal synchronization signal component coupling occurs within each peak frequency of the retained peak frequency group; if such coupling occurs, a camera device is detected. This invention can utilize the specific signal spectrum generated by the coupling modulation of the data signal line and clock signal line on the CSI connection cable during camera operation to detect unknown, concealed cameras. It is suitable for private occasions such as confidential meetings and private conversations, with an effective distance of 20 centimeters.

This invention discloses a Transformer-based human pose estimation method. This method extracts global and local features through a dual-branch network combining CNN and Transformer. Subsequently, a multi-stage regression method of "human center-limb trunk-keypoints" is used to directly regress human keypoints. Compared with traditional detection methods, this invention can simultaneously consider both global and local features, overcoming the shortcomings of CNN ignoring global features and Transformer ignoring local features. Furthermore, the multi-stage regression method adds "limb trunk" intermediate points, reducing the regression difficulty for keypoints far from the human center, improving the accuracy of keypoint regression, and further enhancing the accuracy of human pose estimation.

This invention discloses a method for detecting recording devices based on electromagnetic signals, comprising the following steps: acquiring radio frequency signals in the environment; performing spectrum analysis on the radio frequency signals after filtering, denoising, and smoothing; searching for peak points belonging to the frequency band of the electromagnetic radiation signal of the recording device in the spectrum, and calculating the average peak interval; matching the peak interval with the frequency feature library of the recording device's AD conversion drive signal, and confirming the detection of the recording device if they are equal. This invention can detect unknown and concealed recording devices by utilizing the electromagnetic radiation signals generated by the AD conversion of the recording device during normal operation, and can locate the device by multiple detections from different directions. It is suitable for private occasions such as confidential meetings and private conversations, with an effective distance of 20 centimeters.

The embodiment of the present invention provides a range hood, an air intake control method for the range hood, and a storage medium. The range hood includes a pyroelectric module, a camera, a control device, and an air guide plate assembly. The pyroelectric module is used to detect the temperature of a first burner and a second burner on a stove. The camera is used to collect an image of the stove. The air guide plate assembly includes a first air guide plate, a second air guide plate, a third air guide plate, and a driving mechanism. The driving mechanism is used to drive the first air guide plate, the second air guide plate, and the third air guide plate to open and close the first air inlet, the second air inlet, and the third air inlet of the range hood independently of each other. The third air inlet is located between the first air inlet and the second air inlet. The control device is connected to the pyroelectric module, the camera, and the driving mechanism, respectively, and is used to receive the temperature data collected by the pyroelectric module and the stove image collected by the camera, and control the driving mechanism to drive each air guide plate to open and close based on the temperature data and the stove image. The probability of false detection can be effectively reduced.