41results about How to "Improve detection and recognition efficiency" patented technology

The invention relates to a rapid detection identification method for a specified pedestrian or vehicle in a video monitoring network. According to a target template provided by a user, accurate and rapid detection identification is carried out on a specified target in a monitoring network video. In the method, firstly, normalized scale change is carried out on a target template image and a plurality of kinds of mixed characteristics of the template image are calculated; a mixed Gaussian model is adopted to carry out background modeling on each monitoring video, and then parallel processing is carried out and area filtering and morphological postprocessing are used to extract motion foregrounds; normalized scale change is carried out on all motion foregrounds which meet a preliminary screening condition and a plurality of mixed characteristics are calculated and then weighted similarity distances between the motion foregrounds and the target template are calculated and first N objects, which are smaller than a judgment threshold and have smallest weighted similarity distances with the target template image, are returned and used as detection identification results. On the basis of ensuring the detection identification accuracy, the method significantly improves the processing speed of an algorithm through methods of parallel processing and frame skipping and the like.

The invention discloses a ship target detection and identification method for a satellite-borne remote sensing optical image. The method comprises the following steps of preprocessing, ineffective information removal, connected region marking, feature extraction and classifier design. A coarse-to-fine strategy is adopted, firstly, a satellite image is subjected to downsampling and Gaussian filtering processing, then, land and isolated noisy points are removed from the processed image, a candidate region is quickly extracted and is marked, feature information is extracted after the image is rotated, then, the feature information is put into a pre-trained classifier to be furthered confirmed and analyzed, false alarms are removed, and a real ship target is found. Meanwhile, the invention also realizes a ship target detection and identification system for the satellite-bone remote sensing optical image. By use of the technical scheme, satellite-borne storage and calculation of limited resources can be effectively utilized to improve the stability and the real time of a ship detection algorithm, meanwhile, a false alarm rate is effectively lowered, and accurate data can be provided for maritime search and rescue and ship image positioning in time.

The invention relates to a radar remote sensing application technology analyzing radar observation information through images, in particular to an SAR image target detection and recognition integration method based on a convolutional neural network (CNN). According to the method, the convolutional neural network is utilized to automatically mine and select target features, shallow features and deep features are fused, a detection task and a recognition task of an SAR target can be completed at the same time, and SAR image target detection and recognition integration is realized. Compared with other SAR target detection and recognition algorithms, the method has higher detection and recognition efficiency and higher applicability.

The invention provides a face part identification method and device. The method comprises: obtaining a depth image; extracting image pixel characteristics in the depth image; inputting the image pixel characteristics into a face deep learning model to perform identification and classification; determining whether the classification of the image pixel characteristics matches existing face part labels in the face deep learning model; and if the classification of the image pixel characteristics matches existing face part labels in the face deep learning model, outputting the labels corresponding to the image pixel characteristics. According to the invention, a depth image characteristic extraction method is employed to guarantee extraction accuracy, and a deep learning model is employed to identify image pixel characteristics, and is capable of performing identification and classification on a plurality of face parts once.

The invention provides a transformer substation insulator infrared image detection method based on artificial intelligence. The transformer substation insulator infrared image detection method comprises the following steps: acquiring an infrared image of a transformer substation insulator through an infrared thermal imager; preprocessing the acquired image through an algorithm; performing target label processing on the obtained data set; dividing the data set into a training set and a test set; constructing an infrared image detection model of the improved feature fusion single-shot multi-boxdetector; carrying out training and parameter adjustment of the model by using a training set in the data set; performing target detection on the trained model by using a test set in the data set so as to prove the effectiveness of the model; through the above steps, automatic detection of the infrared image of the transformer substation insulator is achieved. By adjusting the parameters of the model and adding a feature enhancement module, the feature extraction capability of the model for the insulator is improved, and therefore it is ensured that safe and real-time detection is conducted onthe infrared image of the transformer substation insulator.

The invention discloses an automatic detection and defect identification method for a power grid wire clamp. The method comprises the following steps: acquiring a wire clamp picture through an image acquisition module; training the wire clamp picture to obtain a wire clamp type; classifying, deleting and selecting the wire clamp pictures; and identifying and outputting the fault wire clamp. The invention also discloses an automatic detection and defect identification device for the power grid wire clamp. According to the method and the device, image acquisition is carried out on the wire clampon the power transmission line through the image acquisition module; the Faster R-CNN technology is adopted to train the wire clamp pictures, the augmented training technology is adopted, the defectsthat the time is long and the model updating period is slow due to the fact that new wire clamp features need to be retrained are overcome, fault detection is conducted on the wire clamp, the maintenance cost is greatly saved, and a line patrol system is more efficient and intelligent.

The invention relates to an ink jetting printing head correction method, apparatus and system, a computer apparatus and a storage medium. The method comprises the steps of controlling nozzles of an ink jetting printing head to drop ink droplets on a test substrate; detecting and obtaining the number of the ink droplets dripping on the test substrate, and determining the actual using number of thenozzles on the ink jetting printing head according to the number of the ink droplets; determining whether the actual using number of the nozzles is the same as the target using number of the nozzles or not; and if the actual using number of the nozzles is not the same as the target using number of the nozzles, correcting the current actual using number of the nozzles of the ink jetting printing head to be the same as the target using number. The above method ensures that the actual using number of the nozzles of the ink jetting printing head is the same as the target using number, the problemthat the current actual using number of the nozzles of the ink jetting printing head is not the same as the target using number due to person operation negligence, and consequently a device for ink jetting printing preparation is abnormal is solved, and the accuracy of the using number of the nozzles in ink jetting printing is improved.

The invention provides an electrical equipment infrared image real-time detection and diagnosis method based on lightweight deep learning. The electrical equipment infrared image real-time detection and diagnosis method comprises the following steps: S1, acquiring an infrared image of electrical equipment of a transformer substation through an infrared thermal imager; s2, preprocessing the obtained infrared image through an algorithm to form a data set for training; s3, performing target label processing on the obtained normal and fault data sets of the electrical equipment; s4, randomly distributing the processed data set into a training set and a test set; s5, constructing an infrared image real-time detection and diagnosis model of the improved lightweight single-shot multi-box detector; s6, performing parameter adjustment and training of the model by using the divided training set; and S7, carrying out automatic detection and diagnosis on the trained detection and diagnosis model by using the divided test set so as to prove the effectiveness of the detection and diagnosis model. Through the above steps, real-time detection and diagnosis of infrared images of various electrical devices (especially effective schemes can be deployed in limited environments such as embedded devices) are realized.

The invention discloses a thermostat error-prevention visual detection and recognition device comprising a cabinet and an operation platform which is arranged in the middle of the cabinet. The devicealso comprises a visual imaging system, a visual processing system, a detection clamp platform and a detection displayer. The visual imaging system comprises a side camera unit and an upper camera unit, and the side camera unit and the upper camera unit are arranged at different perspectives of the thermostat. The visual processing system is arranged between the visual imaging system and the detection displayer. The invention also discloses a thermostat error-prevention visual detection method. According to the thermostat error-prevention visual detection and recognition device, multiple different detection points are simultaneously photographed through the dual-angle arranged CCD camera units and the small change of the thermostat is recognized so that misrecognition caused by personnel visual error and fatigue can be avoided, the correct recognition rate can be 100%, the defective products can be effectively distinguished, the mark detection time for each product is within 0.5s and thus the efficiency of detection and recognition can be greatly enhanced.

The invention provides a test question detection and recognition method and device, electronic equipment and a medium, which belong to the technical field of network intelligent education. The test question detection and recognition method comprises the following steps of acquiring a target image, generating a first detection result through a model detection algorithm, wherein the first detectionresult comprises a question information detection result and an answer information detection result, recognizing the first detection result by using an OCR model to obtain a first recognition result,wherein the first recognition result comprises a character line recognition result and a formula recognition result, and matching the first recognition result with preset test question template data in a database, and outputting a correction result according to the similarity. According to the method, a model detection algorithm is adopted for the target image, the question information detection result and the answer information detection result are generated, OCR model recognition is conducted on the two detection results, the character line recognition result and the formula recognition result are recognized, the detection and recognition efficiency of graphs and formulas in test questions and answers is improved, and then the correction efficiency is improved.

The invention relates to an oil pumping unit recognition method and system based on target feature training, and the system comprises an image obtaining module which is used for obtaining an infraredpanoramic image and converting the infrared panoramic image into an image input source of a monitoring identification module; a monitoring identification module which is used for processing the acquired infrared image input, detecting the oil pumping unit in the infrared image, and identifying the position and the working form of the oil pumping unit based on deep learning; and a state analysis module which is used for receiving the processing result of the monitoring identification module and continuously analyzing the working form of the oil pumping unit to obtain the working state of the oil pumping unit. According to the invention, through intelligent identification of the oil pumping unit, the position and the working form of the oil pumping unit in the well pattern are dynamically obtained, so that the identification of the working state of the oil pumping unit is more accurate, and a better detection effect is obtained.

The invention discloses a lane line detecting method based on color lump counting. The lane line detecting method comprises the following steps that firstly, a lane deviation pre-warning system obtains YUV images around a vehicle; secondly, the lane deviation pre-warning system extracts color lump characteristic values of the YUV images; then, the lane deviation pre-warning system builds a decision tree model through the extracted color lump characteristic values; next, the lane deviation pre-warning system judges whether the YUV images are in a lane line area or in a non-lane line area through the decision tree model, if the YUV images are in the lane line area, the next step is executed, and if the YUV images are not in the lane line area, the step is executed repeatedly to judge the next YUV image; and finally, the lane deviation pre-warning system identifies a lane line in the lane line area through a visual identification system. According to the lane line detecting method based on color lump counting, more efficient and more accurate lane line identification and detection are achieved.

The invention provides a method and a system for detecting an ink box. The detection method comprises the following steps that the storage information of the ink box control chip is read; the image data of the ink box at the carrier is obtained by shooting, and the ink box placed at the carrier is weighed to obtain the weight of the ink box; it is judged that whether the ink box can be recycled ornot according to the storage information, the image data and the weight of the ink box. According to the detection method and system, the ink box being recycled is accurately detected to be reprocessed and put into production and use, the ink box can be recycled to use, the service life of the ink box is prolonged, the use rate of the ink box is improved, mechanical automatic detection and recognition are achieved, the accuracy rate is high, the detection and recognition efficiency is greatly improved, and manpower and material resources are saved.

The invention discloses a handwritten Mongolian detection and recognition method based on segmentation and deformation LSTM. The detection of handwritten Mongolian in a complex environment is realized by using a segmentation-based arbitrary shape text detector SAST. A RoI Rotate module is used for combining a text detection function and a text recognition function; The extracted text candidate box is taken as an input image, and the text recognition of the input image is realized by using a deformation LSTM in combination with a CTC module. According to the method, the SAST is utilized to more effectively extract the polygon representation of the text in any shape. Meanwhile, the long-range correlation of pixels can be captured, a more reliable segmentation result can be obtained at a time, the contents of the detection stage and the recognition stage of the handwritten Mongolian are connected through the application of the RoI Rotate module, and the training efficiency can be further improved. The recurrent neural network is combined with the deformation LSTM, so that the recognition accuracy can be further improved in the implementation of handwritten Mongolian recognition.