34results about How to "Overcome missed detection" patented technology

Disclosed is a steel ball surface defect detection automatic sorting device. A light source is arranged on a work bench, a left camera mounted on a left support and a right camera mounted on a right support are symmetrically arranged on the left side and the right side of a light source bowl cover, camera lenses of the left camera and the right camera are arranged symmetrically and corresponding to a left observation hole and a right observation hole formed on the light source bowl cover, a vibration feed table and a material transferring channel connected to a discharge hole of the vibration feeding table are arranged behind the light source, the material transferring channel penetrates through the power source bowel cover and the front of the light source, a sorting mechanism and a collecting box placed below the sorting mechanism are arranged in front of the light source, a discharge terminal of the material transferring channel is inserted into a feeding hole end of the sorting mechanism, the left camera and the right camera are respectively connected with a peripheral component interconnection (PCI) card, PCI cards are respectively connected with a computer and a pneumatic unit, and an output end of the pneumatic unit is connected with the sorting mechanism. According to the steel ball surface defect detection automatic sorting device, the detecting process can achieve automation completely, no easily-worn part exists, the service life is long, and defect detection is accurate and efficient.

The invention discloses a strong convergence field identification and velocity correction method based on Doppler weather radar data, comprising the following steps: designing a new coordinate system to convert radially distributed radar data into grid point data, and simultaneously completing high reflectivity Region extraction; design a detection template, so as to obtain the parameter vector describing the convergence strength and position of the convergence point on the basis of quickly locating the convergence point; propose the idea of ​​generating a convergence zone from the convergence point, and then using the direction of the convergence zone to correct the convergence intensity and algorithm to effectively overcome the problem of underestimation or even missed detection of the strong convergence field at a single elevation angle; determine the profile line according to the strongest convergence point at a single elevation angle to obtain a better profile showing the strong convergence field, and obtain the depth information of the strong convergence field . This method can automatically detect the strong convergence field, and overcome certain underestimation and missed detection problems through the speed correction method, and at the same time automatically draw the profile of the strong convergence field and calculate the characteristic parameters of the strong convergence field in the profile to promote convection Recognition and forecasting of weather hazards.

A video detection method for street order events based on deep learning and motion consistency, involving the fields of artificial intelligence and computer vision. The method steps of the present invention are: 1) algorithm framework: 2) target detection; 3) motion consistency calculation; 4) event judgment. Compared with the prior art, the present invention designs the target detection deep learning network, trains the scene recognition model, and calculates the motion information in the scene to analyze its behavior state, and uses the target detection technology in the static video frame and the dynamic video frame in the field of video intelligent analysis. Combining medium and target behavior analysis technology, multi-condition joint determination of events, designing a detection system for out-of-store business events and road-occupied business events, and accurately and quickly complete the automatic detection of events.

The present invention discloses a real-time automatic threshold calculating method for scanning and monitoring a radio signal, which relates to a calculating method in radio monitoring software. According to the real-time automatic threshold calculating method, according to real-time scanning on one frame of real-time scanning frequency domain monitoring data [Kn], a threshold value [T'n] can be quickly calculated through a real-time automatic threshold calculating method. According to the real-time automatic threshold calculating method, a counting automatic threshold calculating principle is adopted, the threshold value is based on certain time (or number of scanning); detection leakage may occur for the signals with short appearing time, and influence is caused for subsequent data processing (such as occupation rate counting). The real-time automatic threshold calculating method provides a signal determining basis for radio real-time scanning monitoring and overcomes the problems of: incapability of considering large signals and small signals by linear threshold, incapability of differentiating signals and noises, and detection leakage which may be caused by automatic threshold counting.

The invention belongs to the technical field of biomedicine and clinical application, and relates to a polypeptide fragment specifically binding to the serum of Babesia patients and its amino acid sequence. A series of polypeptide fragments are obtained through bit analysis, and the polypeptide fragments are prepared by polypeptide synthesis, which can be used to prepare serological immunodiagnostic reagents for babesiosis. It is suitable for improving the specificity and sensitivity of babesiosis diagnosis.

The invention discloses an inner defect detector for the annular workpieces, comprising a feeding unit, a host unit, a detecting unit, a discharging unit, a collecting unit and a processing circuit, wherein the feeding unit is fixed on a first door-shaped frame of the host unit, the detecting unit is mounted under the panel of the host unit, the discharging unit and the collecting unit are fixed a second door-shaped frame of the host unit, the processing circuit is mounted on the base. The detector has high automation degree of detection, namely, the detector can automatically discharge, detect, classify and collect the annular workpieces, therefore improving the detecting efficiency and the safety; the detector has wide application scope, namely the detector can detect the inner secret defects for most of complex annular workpieces; the automation streamline detector is adopted, the detecting speed is rapid and the continuous operation is performed; the detecting accuracy is high, the audio frequency detecting method can overcome the defects in the traditional method such as missing detection, false detection and the like.