47results about How to "Easy to detect and identify" patented technology

The invention provides a method for extracting pulse waveform feature points, and aims to accurately detect a dicrotic wave and a main pulse wave of a pulse wave. The method comprises the steps that (1), an electrocardiogram (ECG) signal and a pulse signal are synchronously collected; (2), baseline drift, power frequency interference and electromyogram (EMG) interference are removed from the collected ECG signal and the collected pulse signal, so as to eliminate noise; (3), the ECG signal is detected and analyzed, ECG feather information is extracted, and the positions of an R wave and a T wave of the ECG signal are accurately judged; and (4), the positions of the main wave, a dicrotic notch and the dicrotic wave of the pulse signal are determined on the basis of the positions of the R wave and the T wave of the ECG signal. A system employing the method comprises a physiological signal data collecting card used for synchronously collecting the ECG signal and the pulse signal, an ECG detecting module used for detecting the ECG signal of a human body, and a pulse detecting module used for detecting the pulse signal of the human body.

The invention provides a light variation anti-forgery element with multiple anti-forgery functions, which is provided with at least one linear photo-polymerization material layer directed at different directions selectively, at least one optical isotropic material layer and at least one metal layer existing at the local part of the anti-forgery element. The anti-forgery element can have a first-line anti-forgery characteristic provided with a hollow-out metal graphic and a second-line anti-forgery characteristic for hidden information and the like identified by a polaroid. On the basis of the anti-forgery characteristics, the anti-forgery element can combine with the light variation effect of color change along with the angle, and the light variation effect of rainbow holography and / or machine-readable identified encoding characteristics, thus realizing organic combination of the first-line, the second-line and a third-line anti-forgery technologies, having the anti-forgery characteristic which is easy for people to recognize and is difficult to forge and higher anti-forgery level, and providing safe and reliable anti-forgery effect for objects provided with the anti-forgery element. The invention also provides an object with the anti-forgery element.

The invention belongs to the technical field of video detection in deep learning. The invention relates to a vehicle detection method, in particular to a rapid vehicle detection method based on windowcharacteristics. According to the method, a vehicle window is used for replacing a vehicle body to serve as a target object for detection, a residual module of a ResNet network and a multi-scale feature extraction method of an SSD algorithm are combined, a network structure of YOLOv3 is used for reference, and a full convolution detection method with only 24 convolution layers is constructed; Under the condition that the traffic flow is large, during batch testing, the average detection precision is close to 100%, the average detection rate reaches 90%, the detection speed reaches 22 milliseconds per frame, real-time detection of vehicles in a road high-definition monitoring video is achieved, the detection rate of the vehicles in the large traffic flow is effectively increased, and the method has important application value.

The invention discloses an image processing method and device, electronic equipment and a readable storage medium. The image processing method includes obtaining an initial image and performing superpixel segmentation processing on the initial image to obtain segmented image blocks; extracting a region of interest from the segmented image blocks based on image features of a target image; and performing super-resolution reconstruction on the region of interest to obtain an optimized image. By combining the superpixel segmentation and the super-resolution reconstruction, after the initial imageis subjected to the superpixel segmentation processing, the region of interest can be extracted correspondingly based on specific features of the desired target image, and the optimized image is obtained by use of the super-resolution reconstruction method, so that the resolution of the image is greatly improved; and thus interference information can be eliminated, corresponding image features canbe highlighted, i.e., the image processing method provided by the application can optimize to obtain an optimized image with higher definition and resolution, thereby facilitating subsequent detection and recognition.

One embodiment facilitates detection and identification of vehicle parts. During operation, the system stores a captured image of a vehicle, wherein the captured image includes a plurality of parts of the vehicle. The system detects a first area of the captured image in which a first part of the vehicle is located. The system identifies, from a plurality of predetermined classes, a class corresponding to the first part in the detected first area, wherein the class indicates a location of the first part in relation to the vehicle. The system generates a result which indicates a list including an insurance claim item and corresponding damages based on the first area, the first part, and the class corresponding to the first part.

The invention discloses a projection welding bolt torque detection device and a detection method thereof and belongs to the field of automobile body production, manufacturing and detection.The device and method aim to solve the problems that an existing detection means wastes time and labor, the labor intensity of detection personnel is increased, detection cost is increased, and a welded projection welding bolt can not be detected.The projection welding bolt torque detection device comprises a clamp connected with the threaded end of a projection welding bolt, a sleeve arranged on the clamp in a sleeving mode, a conical block arranged on the sleeve in the sleeving mode, a ranging wrench matched with the conical block, a locking nut arranged on the sleeve, and a spring arranged on the sleeve, wherein the two ends of the spring are connected with the conical block and the locking nut respectively.The device can detect the welded projection welding bolt and can be matched with different clamps so that torques of various parts can be conveniently detected, and whether the torque of projection welding bolts meets requirements or not can be quickly detected on a large scale.

The invention provides a parathyroid gland intra-operative real-time detection method based on the Raman spectrum technology, which comprises the following steps of: step 1, detecting tissues to be identified by using a Raman spectrometer to obtain spectral data; step 2, converting the spectral data into a .Txt format, carrying out fluorescence background removal, noise reduction, smoothing and normalization processing, and outputting the spectral data into a .xls format; step 3, establishing a Raman spectrum database, and importing the spectrum data in the .xls format into the man spectrum database; step 4, establishing a machine learning module, wherein the machine learning module establishes a differential diagnosis model according to the Raman spectrum database; and step 5, importing the spectral data of the tissue to be identified into the differential diagnosis model for comparison. According to the parathyroid gland intra-operative real-time detection method based on the Raman spectrum technology, parathyroid glands can be detected and recognized objectively, accurately, simply and conveniently in real time.

The invention discloses a page skipping method based on text hiding and a related device. A skipping link is marked on a server by a target hiding character string in advance, the code of the target hiding character string is embedded in the code of a text loaded in an original page, so that when a client displays the original page, the skipping link is invisible, and when page skipping needs to be executed, a client can determine a target carrier of the skipping link according to the target hiding character string in the original page and can further obtain the hidden target skipping link indicated by the target carrier through the analysis of the server, and thus the client can realize the page skipping according to the target skipping link. Therefore, the skipping link is hidden in the original page by matching and embedding operations of the hiding character string, and thus the skipping link can be prevented from being counterfeited, deleted and modified; and the hiding manner is not limited by the text format or the text statement, thereby being easy to detect and identify; and the correct execution of the skipping operation can be guaranteed.

The invention relates to the technical field of holographic imaging security check, in particular to a three-dimensional holographic imaging security check radar, and particularly provides a three-dimensional holographic imaging security check radar image foreign matter detection method. The foreign matter detection method comprises the following steps: acquiring target detection data of a three-dimensional radar image; inputting the three-dimensional radar image target detection data into a trained convergent anchor-point-free neural network model, and obtaining a three-dimensional radar image foreign matter detection result output by the anchor-point-free neural network model. According to the method, the number of prediction boxes extracted by the neural network model can be greatly reduced, the calculation complexity of the model is reduced, the finally predicted detection box has good flexibility, and the recognition effect of an object with large scale change can be improved; meanwhile, through a multi-level three-dimensional radar image feature extraction mode, the neural network model can obtain more sufficient semantic information, and detection and recognition of objectsin an image scene are facilitated.

The invention discloses a driver posture detection method based on the centroid coordinates of an arm skin color region. Firstly, an image of the driver when driving is collected, and the centroid coordinates of the driver's hands and / or arms are extracted. Secondly, the extracted coordinates and the driver posture classes are used as training samples, and a driver posture detection model is obtained by a training classifier. During the detection, the centroid coordinates of the driver's hands and / or arms in the image to be detected are extracted as the input of the driver posture detection model, and the driver's driving posture is detected by classification and recognition. This method can quickly establish and train the driver posture detection model, so as to achieve the rapid detection of driver posture.

The invention provides a QR code region detection method for improving background prior and foreground prior. The QR code region detection method comprises the following steps: performing convex hulldetection on an input original image G; carrying out super-pixel segmentation; respectively carrying out multi-feature extraction on the image under each super-pixel scale; calculating a background saliency map of the original image G; calculating a foreground saliency map of the original image G, fusing the obtained final background saliency map and the final foreground saliency map to obtain a weak saliency map, training by adopting a multi-kernel learning enhancement method according to a training sample generated by the weak saliency map to obtain a strong saliency model, and applying themodel to all test samples to obtain a strong saliency map; and finally, weighting and fusing the strong saliency map and the weak saliency map to obtain a final saliency map, wherein the highlighted part in the final saliency map is the QR code area in the image. According to the QR code region detection method, the salient targets can be highlighted accurately and consistently, and the QR code area in the image can be detected accurately. The QR code region detection method provided by the invention has more advantages in the aspect of accuracy of salient target detection.

The invention relates to a power transformation equipment defect detection method based on sensitive position dependence analysis. Compared with the prior art, the defect that the power transformation equipment defect detection rate is low is overcome. The method comprises the following steps: acquiring a defect image set of the power transformation equipment; constructing a defect detection network; training a defect detection network; acquiring a defect image to be detected; and detecting defects of substation equipment. The relevance between the pixels with high relevance in the power transformation equipment image is fully considered, so that the recombined target features are not only pure target information, the relevance contains the image background and the target context dependency information, and detection and recognition are more facilitated.

The invention discloses an method for extinguishing and cooling activated carbon in a front discharge hopper of an adsorption tower. The method comprises the following steps: 1) acquiring a thermal imaging image of a material entering a tail imaging area (3) of a vibrating screen (2) by using a thermal imager (1); 2) judging whether the material entering the imaging area (3) has a high-temperature point or not according to the thermal imaging image, and if it is judged that the thermal imaging image has the high-temperature point, recording the finding position of the material at the high-temperature point in the tail imaging area (3) of the vibrating screen (2); and (3) when the materials at the high-temperature point move into a discharge hopper (5) connected with a discharge port of a conveyor (4), carrying out fire extinguishing and cooling treatment on the high-temperature materials through a spontaneous combustion activated carbon extinguishing and cooling device arranged at the upper part of the discharge hopper (5). High-temperature activated carbon is detected at the tail part of the vibrating screen, and is treated in a flowing and falling state when the high-temperature activated carbon falls into the stock bin, so the problem that high-temperature activated carbon particles are difficult to comprehensively detect and treat is solved, and the safety of the system is improved.

The invention provides a method for detecting and identifying dense weak and small targets in a wide remote sensing image. According to the method, sample amplification is carried out by means of Mosaic, rotation, MixUp, zooming and the like, and abundant and multi-scale training samples are obtained; secondly, improving a YOLOv5 network by utilizing Transform, adding a weak and small target detection head, enhancing the characterization capability of the model on global information and context information in an image, and improving the weak and small target detection and recognition performance in a dense scene; meanwhile, a multi-model fusion prediction method is used, the prediction precision is improved, and the model is prevented from falling into local optimum; and finally, detecting and identifying dense weak and small targets in the wide remote sensing image by using a block detection method. According to the method, multi-model fusion prediction is used, the model target detection and identification capability can be improved, and the model is prevented from falling into local optimum.

The invention discloses ultra-low color difference concrete based on heavy calcium carbonate powder. The ultra-low color difference concrete comprises the following components in percentage by weight: 200-450 kg / m < 3 > of cement, 0-150 kg / m < 3 > of slag powder, 1000-1200 kg / m < 3 > of gravel, 700-800 kg / m < 3 > of sand, 150-180 kg / m < 3 > of liquid water, 0-80 kg / m < 3 > of solid ice, 70-150 kg / m < 3 > of heavy calcium carbonate powder and 0.8-5.0 kg / m < 3 > of a special additive. The invention further discloses a preparation method of the ultra-low color difference concrete based on the heavy calcium powder. Through reasonable proportion design, the poured concrete is uniform in appearance color, free of bubbles and very good in facing effect, the effect of bare concrete can be achieved without modification, the facing effect of a bridge structure is met, and high-added-value utilization is provided for heavy calcium carbonate powder and granulated blast furnace slag powder solid waste; the cement consumption is reduced, so that carbon emission is reduced, and low-carbon concrete is prepared; in a three-phase system of heavy calcium powder, granulated blast-furnace slag powder and cement, the machine-made sand can be applied to bridge facing concrete with higher concrete appearance requirements, and the application range of the machine-made sand is further widened.

The invention relates to the technical field of construction management, in particular to an intelligent construction site management system and method based on Internet +, the method is applied to the system, and the system comprises a security terminal, an acousto-optic early warning terminal, a plurality of cameras and a server. The server comprises a video monitoring module which is used for obtaining real-time monitoring data of a construction site through a plurality of cameras arranged on the construction site; a data storage module which is used for storing real-time monitoring data; an anomaly detection module used for detecting and analyzing figure images and vehicle images appearing in the real-time monitoring data and generating early warning information; an acousto-optic early warning module used for generating early warning information, sending a voice early warning prompt through an acousto-optic early warning terminal, and sending the voice early warning prompt to the security terminal; and an exception handling module used for handling the exception behavior. The problem that the loss cannot be stopped in time due to the fact that the construction site building materials of the construction site construction party cannot be stopped when being stolen is solved.

The invention provides an edgewise conveying device and a control method thereof. The edgewise conveying device comprises a frame mechanism, a lifting mechanism and an edgewise acting mechanism. The frame mechanism comprises a plurality of cross beams and a plurality of longitudinal beams, and the two ends of each cross beam are connected between the corresponding longitudinal beams. One side of the frame mechanism is hinged to a conveying support frame on one side. The lifting mechanism comprises a lifting hydraulic cylinder and a mounting base. The mounting base is fixed to a working plane.The lifting hydraulic cylinder comprises a lifting cylinder body and a lifting cylinder rod. The lifting cylinder body is hinged to the mounting base. One end of the lifting cylinder rod is hinged tothe other side of the frame mechanism. The lifting hydraulic cylinder can drive the frame mechanism to rotate around the conveying support frame on one side. The edgewise acting mechanism comprises atransverse moving hydraulic cylinder and a transverse moving rod. The transverse moving hydraulic cylinder comprises a transverse moving cylinder body and a transverse moving cylinder rod. The transverse moving rod is connected with the transverse moving cylinder rod. The transverse moving hydraulic cylinder can drive the transverse moving rod to push goods on a conveying platform to one side of the conveying platform.

The invention belongs to the technical field of imaging spectrums, and particularly relates to a double-light-source detection and recognition method, which comprises a first light source and a second light source, and comprises the following steps: arranging the first light source on one side of a detection table, arranging the second light source on the other side of the detection table, and moving an object to be detected to the center of the detection table; starting the first light source and the second light source to irradiate the to-be-detected object; irradiating the surface of a to-be-detected object through a first light source, collecting reflected light of the to-be-detected object irradiated by the first light source through an output light path, and inputting the reflected light into a spectrograph; if the to-be-detected object is a transparent object, irradiating the back surface of the to-be-detected object through a second light source, collecting transmission light after the to-be-detected object is irradiated by the second light source through an output light path, and inputting the transmission light into a spectrograph; according to the spectral data of the to-be-detected object, the material information of the to-be-detected object is analyzed and obtained, and a user can simultaneously detect and identify transparent and non-transparent objects without repeatedly adjusting the building mode of the light path.

The invention provides a method for conducting secondary fire extinguishing and cooling treatment on high-temperature activated carbon. The method comprises the steps that firstly, when it is detected that materials entering a vibrating screen (1) have high-temperature points, primary fire extinguishing and cooling treatment is conducted on the high-temperature materials through a first spontaneous combustion activated carbon extinguishing and cooling device arranged on a discharging pipe (3) between the vibrating screen (1) and a conveyor (2); and secondly, when the materials at the high-temperature point move into a discharge hopper (4) connected with a discharge port of the conveyor (2), secondary fire extinguishing and cooling treatment is conducted on the high-temperature materials through a second spontaneous combustion activated carbon extinguishing and cooling device arranged at the upper part of the discharge hopper (4). High-temperature activated carbon at the tail of the vibrating screen can be treated in time through primary treatment of the discharging pipe, the high-temperature activated carbon can be further extinguished and cooled through secondary treatment of the discharging hopper, and therefore the extinguishment and cooling effects of high-temperature materials are ensured, and the safety of the system is improved.

The invention belongs to the field of machine learning, and particularly relates to a multi-scale widened residual network, a small target recognition and detection network and an optimization methodthereof. The multi-scale widened residual network comprises a multi-scale widened convolutional layer and a multi-scale widened residual network unit structure which are connected in series; the multi-scale widened convolution layer comprises a plurality of convolution kernels of different scales which are arranged in parallel, and the output of the multi-scale widened convolution layer is the combination of the extracted features of the convolution kernels of different scales; the multi-scale widening residual network unit structure comprises a plurality of multi-scale widening convolution layers which are arranged in series, and skip layer connection between the two multi-scale widening convolution layers is set, so that output features of the two multi-scale widening convolution layersare directly subjected to maximum fusion. Small target recognition detection is carried out based on the obtained features, and the accuracy of small target object recognition can be improved.

A quick and automatic black and odorous water body identification system comprises a box body, a movable shell, two groups of water pipes and a plurality of groups of storage boxes; the interior of the box body is divided into an upper-layer first mounting bin and a lower-layer second mounting bin; a guide sliding assembly slidably connected with the moving shell and a placement table are arranged in the second mounting bin. A liquid guide pipe and a liquid pumping device are arranged in the movable shell; a flow detection module and two groups of electrode modules are arranged on the liquid guide pipe; electrode ends of the two groups of electrode modules extend into the liquid guide pipe in a sealing manner; the two groups of electrode modules are electrically connected with positive and negative electrodes of the oxidation-reduction potential detection module respectively; the liquid guide pipe is connected with the liquid pumping device, and the liquid guide pipe and the liquid pumping device are respectively connected with two groups of water pipes; temperature sensors are arranged on the water pipes; a heating assembly and a cooling assembly are arranged on the placing table; a baffle is arranged in the first mounting bin and forms an enclosed mounting bin, and a touch display screen is arranged on the baffle; and a controller, a power supply module and an oxidation-reduction potential detection module are arranged in the mounting bin. The black water body can be conveniently and rapidly detected and identified by the quick and automatic black and odorous water body identification system.

The invention belongs to the technical field of water supply network leakage detection, and provides a water supply network leakage identification method based on deep learning in order to solve the problems that leakage characteristics are difficult to extract, parameters are sensitive and the like in a conventional noisemeter processing method, and the method comprises the steps that firstly, an audio signal is converted into a frequency spectrum thermodynamic diagram, and time-frequency domain information of an original signal is fully reserved; and secondly, a large number of existing actual samples are utilized, features are automatically extracted through the self-learning ability of the CNN model for recognition, the limitation of human experience is broken through, and compared with a conventional signal processing method, the method has more accurate leak detection ability and stronger generalization characteristics.