WO2020164282A1 - Procédé et appareil de reconnaissance de cible d'image basée sur yolo, dispositif électronique et support de stockage - Google Patents

Procédé et appareil de reconnaissance de cible d'image basée sur yolo, dispositif électronique et support de stockage Download PDF

Info

Publication number
WO2020164282A1
WO2020164282A1 PCT/CN2019/118499 CN2019118499W WO2020164282A1 WO 2020164282 A1 WO2020164282 A1 WO 2020164282A1 CN 2019118499 W CN2019118499 W CN 2019118499W WO 2020164282 A1 WO2020164282 A1 WO 2020164282A1
Authority
WO
WIPO (PCT)
Prior art keywords
detection frame
image
classification
yolo
preset
Prior art date
Application number
PCT/CN2019/118499
Other languages
English (en)
Chinese (zh)
Inventor
赵峰
王健宗
肖京
Original Assignee
平安科技（深圳）有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 平安科技（深圳）有限公司 filed Critical 平安科技（深圳）有限公司
Publication of WO2020164282A1 publication Critical patent/WO2020164282A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]

Definitions

  • This application relates to the field of computer learning and image recognition, and more specifically, to a YOLO-based image target recognition method, device, electronic equipment and storage medium.
  • YOLO solves object detection as a regression problem, based on a single end-to-end network, to complete the input from the original image to the output of the object position and category.
  • YOLO The core idea of YOLO is to use the entire graph as the input of the network, and directly return to the position of the bounding box and the category to which the bounding box belongs in the output layer. The inventor realized that based on the use of YOLO's high-speed operation, how to design a method that can improve the accuracy of YOLO is an urgent need to solve at present.
  • this application proposes a YOLO-based image target recognition method, device, electronic equipment and storage medium.
  • the technical solution of the present application provides a method for image target recognition based on YOLO, including:
  • the detection frame and the classification identification information are used as the classification result of the identification.
  • the technical solution of the present application also proposes an image target recognition device based on YOLO, which includes: an input module to receive the image to be detected;
  • the adjustment module adjusts the size of the image to be detected received by the input module according to preset requirements, and generates the first detected image
  • the matching recognition module sends the first detection image generated by the adjustment module to the neural network model for matching recognition, and generates a detection frame, classification identification information, and classification probability values corresponding to the classification identification information;
  • a judging module judging whether the classification probability value is greater than a preset classification probability threshold, if it is not greater, send a signal to the matching recognition module, if it is greater, send a signal to the classification module;
  • the classification module uses the detection frame and the classification identification information as the classification result of the identification.
  • the technical solution of the present application also proposes an electronic device, including a memory, a processor, and a camera device.
  • the memory includes a YOLO-based image target recognition program, and the YOLO-based image target recognition program is used by the processor.
  • the steps of the above-mentioned YOLO-based image target recognition method are realized during execution.
  • the fourth aspect of the present application also provides a computer non-volatile readable storage medium, the computer non-volatile readable storage medium includes a YOLO-based image target recognition program, the YOLO-based image target recognition program is When the processor is executed, the steps of the above-mentioned YOLO-based image target recognition method are realized.
  • This application proposes a YOLO-based image target recognition method, device, system and storage medium.
  • This method judges the classification and recognition probability, and only uses the recognition information as the recognition result when the preset classification probability threshold is reached, which improves the accuracy of image recognition and the sense of recognition experience.
  • the present application can also adjust the position of the detection frame in real time, which effectively improves the detection efficiency and accuracy; the detection time is reduced by optimizing the detection calculation method. After experiments and verifications, the method of this application is superior to the detection method of the prior art. More embodied in improved recognition accuracy and increased computing speed.
  • Figure 1 is a flow chart of an image target recognition method based on YOLO in this application
  • Figure 2 shows a schematic diagram of the convolution operation in the classification process of this application
  • Figure 3 shows a block diagram of an electronic device of the present application
  • Fig. 4 shows a schematic diagram of a specific embodiment of the present application.
  • Fig. 1 is a flow chart of an image target recognition method based on YOLO in this application.
  • the technical solution of the present application provides a YOLO-based image target recognition method, including:
  • S104 Adjust the size of the image to be detected according to a preset requirement to generate a first detected image
  • S106 Send the first detection image to a neural network model for matching recognition, and generate a detection frame, classification identification information, and classification probability values corresponding to the classification identification information;
  • the size is the size specified by the neural network model.
  • the size of the generally selected image will be smaller than the size of the image to be detected, which can ensure the speed of calculation processing and can quickly perform class recognition.
  • 448*448 or 416*416 is selected.
  • the size selection in this step can be set according to actual needs, and is not limited to the above-mentioned sizes and cannot limit the protection of this application. range.
  • the first detection image is sent to a neural network model to generate a detection frame, classification identification information, and classification probability values corresponding to the classification identification information.
  • a neural network model to generate a detection frame, classification identification information, and classification probability values corresponding to the classification identification information.
  • the classification probability threshold is set to 90%, when detecting a picture containing a kitten, if the probability of identifying the kitten in the detection frame exceeds 90% , It means that a kitten is circled in the detection box and the cat in the picture has been identified.
  • the classification probability value is less than the preset classification probability threshold, it will return to step S106 for re-identification until the classification probability value is greater than the preset classification probability threshold.
  • the neural network model performs multi-layer convolution operations on the image.
  • the described YOLO convolution operation is a conventional operation in the field, and belongs to the prior art, and will not be repeated in this application.
  • step S106 it includes after step S106:
  • the step After the step of generating the detection frame, the classification identification information, and the classification probability value corresponding to the classification identification information, the step includes:
  • the remaining detection frames of the same type are calculated for the coincidence degree, and the detection frame with the highest coincidence degree is retained.
  • the method before receiving the image to be detected in the step S102, the method further includes:
  • the neural network model is trained through the following steps:
  • the preprocessed image set is trained to obtain a neural network model with an input interface and an output interface.
  • the step of obtaining the training image data set includes:
  • a set number of positive samples and negative samples of each tag in the total set of identification tags are selected from the picture library to form the training set and the validation set, where a positive sample of a label is a picture containing the object corresponding to the label, and a negative of a label
  • the sample is a picture that does not contain the object corresponding to the label
  • the training set is the image data of the positive sample and the negative sample
  • the verification set is the label sequence of the positive sample and the negative sample
  • the output of the neural network model Is the predicted label sequence of the samples in the training set.
  • the training image data set has 1,000 object categories and 1.2 million training images.
  • the preprocessing includes one or more of rotation, contrast enhancement, tilt, and scaling.
  • the image will be distorted to a certain extent.
  • the training of the distorted image can be Increase the accuracy of the final image recognition.
  • the step of generating the detection frame is specifically as follows:
  • Predict the dynamic detection frame perform iterative prediction on the generated detection frame, and generate the latest detection frame
  • the latest detection frame coincidence degree is greater than or equal to the preset coincidence degree threshold, keep the latest detection frame; if the latest detection frame coincidence degree is less than the preset coincidence degree threshold, continue to predict the dynamic detection frame ;
  • the initial preset coordinate point is the coordinate point of the preset detection frame, which may be automatically generated during training and recognition detection, or may be generated by a person skilled in the art according to actual needs.
  • the prediction of the dynamic detection frame, the iterative prediction of the generated detection frame, and the generation of the latest detection frame are specifically as follows:
  • b x , b y , b w , and b h are the four coordinate point values of the latest detection frame. It should be noted that the detection frame is a quadrilateral, and the position of the quadrilateral detection frame can be determined by the values of 4 points.
  • the network predicts the 4 coordinates of each detection frame, (t x , t y , t w , t h ). If the cell deviates from the coordinates (c x , c y ) of the upper left corner of the image, the coordinates of the latest detection frame expressed by the above formula can be obtained.
  • each box uses multi-label classification to predict the classes that the bounding box may contain.
  • this application uses binary cross entropy loss technology for class prediction.
  • the main purpose of using binary cross entropy loss for category prediction is that the applicant finds that the softmax technology does not require good performance, but only uses an independent logical classifier, so this step does not need to use the softmax technology.
  • the binary cross entropy loss technology will provide more help.
  • the binary cross-entropy loss technology is a common technology in the field, and those skilled in the art can implement the binary cross-entropy loss technology according to requirements, and this application will not repeat them one by one.
  • the convolution operation of each layer is calculated by alternating 3 ⁇ 3 and 1 ⁇ 1 convolution layers.
  • the applicant has found through a limited number of actual tests that the use of the above-mentioned convolutional layer for alternating operation can increase the accuracy and effectively increase the operation speed.
  • the alternate calculation of the convolutional layer is specifically as follows: firstly, a 3 ⁇ 3 convolution operation is used, and then a 1 ⁇ 1 convolution operation is used, and the operations are alternately performed in turn until all the convolutional layers have participated in the operation.
  • the size is the size specified by the neural network model.
  • the classification probability value in the detection frame is calculated, and the optimal N detection frames of the same type are selected. It should be noted that the size of the detection frame is dynamically predicted, and the process of dynamic prediction is the solution described above.
  • the probability threshold is used to filter the M classification probability values of all the detection frames, and a set of screening rules is formulated:
  • Calculate the classification probability value of each detection frame arrange its classification probability values from largest to smallest, and select the highest ranked category. It can be said that this step is the first round of screening.
  • the M categories of each detection frame are evaluated first, and a champion category with the highest probability value is selected.
  • the highest ranked category is compared with a preset probability threshold. If it is greater than or equal to the preset probability threshold, the detection frame is retained; if it is less than the preset probability threshold, the detection frame is deleted. It can be said that in the second round of screening, the champion classification is compared with the probability threshold, and the check box with a value greater than the probability threshold is eligible to enter the final.
  • the probability threshold can be set to 0.24 (24%). By comparison, the preset detection frame is displayed on the picture, and it can be seen that as long as the classification probability value is greater than or equal to the 0.24 probability threshold, it can be displayed.
  • the overlap degree calculation is performed on the N detection frames of the same type, and the detection frame with the highest overlap degree is retained.
  • the coincidence degree calculation (IOU) is performed in pairs. If the coincidence degree calculation value IoU>0.3, the detection frame with low probability is eliminated.
  • Figure 2 shows a schematic diagram of the convolution operation in the classification process of this application.
  • the neural network model adopts 53 layers of convolution operation, and the convolution operation of each layer is 3 ⁇ 3 and 1 ⁇ 1 convolution layers alternately.
  • This feature extraction method realizes the highest measurement floating point operation per second. This also means that the neural network structure can make better use of the machine's GPU, improve the evaluation efficiency, and thus increase the speed. Because the ResNets technology has too many levels and is not efficient, the convolution operation described in this application can have higher efficiency and higher accuracy.
  • each neural network is trained with the same settings and tested with a single cropping accuracy of 256 ⁇ 256.
  • the performance of the classifier using the feature extraction of the present application is comparable to the most advanced classifier in the prior art, but there are fewer floating point operations and faster speed.
  • FIG. 3 shows a block diagram of the application of the above-mentioned YOLO-based image target recognition method of this application to an electronic device.
  • the technical solution of the present application also proposes an electronic device 2, which includes a memory 201, a processor 202, and a camera 203.
  • the memory 201 includes an image target recognition program based on YOLO.
  • the YOLO image target recognition program is executed by the processor, the following steps are implemented:
  • the detection frame and the classification identification information are used as the classification result of the identification.
  • the first detection image is sent to a neural network model to generate a detection frame, classification identification information, and classification probability values corresponding to the classification identification information.
  • a neural network model to generate a detection frame, classification identification information, and classification probability values corresponding to the classification identification information.
  • the classification probability threshold is set to 90%, when detecting a picture containing a kitten, if the probability of identifying the kitten in the detection frame exceeds 90% , It means that a kitten is circled in the detection box and the cat in the picture has been identified.
  • the classification probability value is less than the preset classification probability threshold, it will return to step S106 for re-identification until the classification probability value is greater than the preset classification probability threshold.
  • the neural network model performs multi-layer convolution operations on the image.
  • the described YOLO convolution operation is a conventional operation in the field, and belongs to the prior art, and will not be repeated in this application.
  • the method before the receiving the image to be detected, the method further includes:
  • the neural network model is trained through the following steps:
  • the preprocessed image set is trained to obtain a neural network model with an input interface and an output interface.
  • the step of generating the detection frame is specifically as follows:
  • the latest detection frame coincidence degree is greater than or equal to the preset coincidence degree threshold, keep the latest detection frame; if the latest detection frame coincidence degree is less than the preset coincidence degree threshold, continue to predict the dynamic detection frame ;
  • the prediction of the dynamic detection frame, the iterative prediction of the generated detection frame, and the generation of the latest detection frame are specifically as follows:
  • dimensional clustering can be used as an anchor frame to dynamically predict the detection frame, and the detection frame is also a bounding box.
  • the network predicts the 4 coordinates of each detection frame, t x , t y , t w , and t h . If the cell deviates from the upper left corner of the image (c x , c y ), the coordinates of the latest detection frame expressed by the above formula can be obtained, where b x , b y , b w , and b h are the coordinates of the latest detection frame.
  • Four coordinate point values It should be noted that the detection frame is a quadrilateral, and the position of the quadrilateral detection frame can be determined by the values of 4 points.
  • each box uses multi-label classification to predict the classes that the bounding box may contain.
  • this application uses binary cross-entropy loss technology for class prediction.
  • the main purpose of using binary cross entropy loss for category prediction is that the applicant finds that the softmax technology does not require good performance, but only uses an independent logical classifier, so this step does not need to use the softmax technology.
  • the binary cross entropy loss technology will provide more help.
  • the binary cross-entropy loss technology is a common technology in the field, and those skilled in the art can implement the binary cross-entropy loss technology according to requirements, and this application will not repeat them one by one.
  • the method before the receiving the image to be detected, the method further includes:
  • Image training is performed to obtain a neural network model; the neural network model is a model with an input interface and an output interface obtained by image training for different types of pictures.
  • the convolution operation of each layer is calculated by alternating 3 ⁇ 3 and 1 ⁇ 1 convolution layers.
  • the applicant has found through a limited number of actual tests that the use of the above-mentioned convolutional layer for alternating operation can increase the accuracy and effectively increase the operation speed.
  • the alternate calculation of the convolutional layer is specifically as follows: firstly, a 3 ⁇ 3 convolution operation is used, and then a 1 ⁇ 1 convolution operation is used, and the operations are alternated in turn until all the convolutional layers have participated in the operation.
  • the size is the size specified by the neural network model.
  • This feature extraction method realizes the highest measurement floating point operation per second. This also means that the neural network structure can make better use of the machine's GPU, improve the evaluation efficiency, and thus increase the speed. Because the ResNets technology has too many levels and is not efficient, the convolution operation described in this application can have higher efficiency and higher accuracy.
  • each neural network is trained with the same settings and tested with a single cropping accuracy of 256 ⁇ 256.
  • the performance of the classifier using the feature extraction of the present application is comparable to the most advanced classifier in the prior art, but there are fewer floating point operations and faster speed.
  • the probability threshold is used to filter the M classification probability values of all the detection frames, and a set of screening rules is formulated:
  • Calculate the classification probability value of each detection frame arrange its classification probability values from largest to smallest, and select the highest ranked category. It can be said that this step is the first round of screening.
  • the M categories of each detection frame are evaluated first, and a champion category with the highest probability value is selected.
  • the highest ranked category is compared with a preset probability threshold. If it is greater than or equal to the preset probability threshold, the detection frame is retained; if it is less than the preset probability threshold, the detection frame is deleted. It can be said that in the second round of screening, the champion classification is compared with the probability threshold, and the check box with a value greater than the probability threshold is eligible to enter the final.
  • the probability threshold can be set to 0.24 (24%). By comparison, the detection frame that passed the preliminaries is displayed on the picture. It can be seen that as long as the classification probability value is greater than or equal to the 0.24 probability threshold, it can be displayed.
  • the overlap degree calculation is performed on the N detection frames of the same type, and the detection frame with the highest overlap degree is retained.
  • the coincidence degree calculation (IOU) is performed in pairs. If the coincidence degree calculation value IoU>0.3, the detection frame with low probability is eliminated.
  • this application also proposes an image target recognition device based on YOLO, including: an input module to receive the image to be detected;
  • the adjustment module adjusts the size of the image to be detected received by the input module according to preset requirements, and generates the first detected image
  • the matching recognition module sends the first detection image generated by the adjustment module to the neural network model for matching recognition, and generates a detection frame, classification identification information, and classification probability values corresponding to the classification identification information;
  • a judging module judging whether the classification probability value is greater than a preset classification probability threshold, if it is not greater, send a signal to the matching recognition module, if it is greater, send a signal to the classification module;
  • the classification module uses the detection frame and the classification identification information as the classification result of the identification.
  • it further includes a training module to perform image training to obtain a neural network model, and the training module includes:
  • Data set acquisition unit to acquire training image data set
  • a preprocessing unit which performs image preprocessing on the training image data set to obtain a preprocessed image set
  • the training unit trains the preprocessed image set to obtain a neural network model with an input interface and an output interface.
  • the aforementioned data set obtaining unit includes:
  • Tag library which stores different tags and tag sequences corresponding to different objects
  • Picture library which stores the image data and label sequence of pictures
  • the screening unit selects a set number of positive samples and negative samples of each tag in the total identification tag set from the picture library to form a training set and a validation set, where a positive sample of one label is a picture containing the object corresponding to the label, and one The negative sample of the label is a picture that does not contain the object corresponding to the label, the training set is the image data of the positive sample and the negative sample, the verification set is the label sequence of the positive sample and the negative sample, the neural network The output of the model is the predicted label sequence of the samples in the training set.
  • the above-mentioned matching recognition module includes:
  • the initial detection frame generating unit generates the initial detection frame according to the initial preset coordinate points
  • the prediction unit predicts the dynamic detection frame, iteratively predicts the generated detection frame, and generates the latest detection frame;
  • the coincidence degree obtaining unit calculates the coincidence degree of the latest detection frame
  • the screening unit if the latest detection frame coincidence degree is greater than or equal to the preset coincidence degree threshold, keep the latest detection frame; if the latest detection frame coincidence degree is less than the preset coincidence degree threshold, send a signal to the prediction unit Then continue to predict the dynamic detection frame;
  • the detection frame generation unit generates N detection frames of the same category.
  • the above prediction unit includes:
  • Prediction sub-unit predict the 4 coordinates of each detection frame, (t x , t y , t w , t h );
  • Update the subunit, by predicting the width p w and height p h of the detection frame predicted by the subunit, and update the coordinates of the detection frame, the coordinates of the latest detection frame are:
  • b x , b y , b w , and b h are the four coordinate point values of the latest detection frame.
  • 53 layers of convolution operation are used in the neural network model, and the convolution operation of each layer is alternately calculated by 3 ⁇ 3 and 1 ⁇ 1 convolution layers.
  • the judgment module includes:
  • the classification probability obtaining unit calculates the classification probability value of each detection frame
  • the first screening unit arranges the classification probability value of each detection frame from largest to smallest, and selects the highest ranked category;
  • the second screening unit compares the highest ranked category with a preset probability threshold, and if it is greater than or equal to the preset probability threshold, then keep the detection frame; if it is less than the preset probability threshold, delete the Check box,
  • the classification module includes a third screening unit, which calculates the degree of coincidence of the reserved detection frames of the same type, retains the detection frame with the highest degree of coincidence, and uses the detection frame with the highest degree of coincidence and its corresponding classification identification information as the recognized classification result.
  • Figure 4 shows a schematic diagram of an embodiment of the present application.
  • the number of convolutional layers in the neural network model is set to 0-52. Then receive the first detected image after size adjustment.
  • the size of the first detected image is 416*416.
  • the specific size can be set according to actual computing requirements and computing capabilities.
  • 416*416 is selected for description, and the color is Color photo.
  • the 0th layer of the neural network model receives the 416*416 size, 3-channel (RGB) color first detection image, and performs the convolution operation.
  • the 52nd layer performs a convolution operation on the feature picture, and the final output one-dimensional prediction array contains 13*13*5*85 values. Reduce the multi-dimensional array or matrix to a one-dimensional array through a series of operations.
  • the one-dimensional array is the prediction array.
  • the number 13*13 in the 13*13*5*85 values represents the width*height of the feature map, and there are a total of 13*13 feature units.
  • YOLO divides the original picture (416*416) into 13*13 cells on average, and each feature unit corresponds to a picture area.
  • the specific size can be set by those skilled in the art according to actual computing requirements and computing capabilities.
  • Number 5 Represents 5 bounding boxes with different shapes. YOLO will generate 5 bounding boxes in each image area, and use the center of the area as the center of the detection box to detect objects, so YOLO will use 13*13*5 detection frames to detect a picture or image.
  • Each detection frame contains 4 coordinate values (x, y, width, height)
  • Each detection frame has a confidence value of the detected object, which is also the above-mentioned confidence (0 ⁇ 1), which is understood as the confidence probability of detecting the object, that is, the confidence value.
  • Each detection frame has 80 classification detection probability values (0 ⁇ 1), which means that the objects in the detection frame may be the probability of each classification respectively.
  • the above process is to divide a 416*416 picture into 13*13 picture areas.
  • Each picture area generates 5 detection frames, and each detection frame contains 85 values (4 coordinates).
  • the final one-dimensional prediction array (predictions) represents the detected objects in the picture, the array contains a total of 13*13*5*85 numerical predictions[0 ] ⁇ predictions[13*13*5*85-1].
  • this application also proposes a computer-readable storage medium including a YOLO-based image target recognition program, which, when executed by a processor, implements the steps of the above-mentioned YOLO-based image target recognition method.
  • This application proposes a YOLO-based image target recognition method, device, electronic equipment and storage medium. This method can effectively improve the detection accuracy and reduce the detection time. After experiments and verifications, the method of this application is superior to the detection method of the prior art. More embodied in improved recognition accuracy and increased computing speed.
  • the disclosed device and method may be implemented in other ways.
  • the device embodiments described above are merely illustrative.
  • the division of the units is only a logical function division, and there may be other divisions in actual implementation, such as: multiple units or components can be combined, or It can be integrated into another system, or some features can be ignored or not implemented.
  • the coupling, or direct coupling, or communication connection between the components shown or discussed may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical, mechanical or other forms. of.
  • the units described above as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • the functional units in the embodiments of the present application can all be integrated into one processing unit, or each unit can be individually used as a unit, or two or more units can be integrated into one unit;
  • the unit can be implemented in the form of hardware, or in the form of hardware plus software functional units.
  • the foregoing program can be stored in a computer readable storage medium.
  • the execution includes The steps of the foregoing method embodiment; and the foregoing storage medium includes: removable storage devices, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks or optical disks, etc.
  • the medium storing the program code.
  • the above-mentioned integrated unit of this application is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer readable storage medium.
  • the computer software product is stored in a storage medium and includes several instructions for A computer device (which may be a personal computer, a server, or a network device, etc.) executes all or part of the methods described in the various embodiments of the present application.
  • the aforementioned storage media include: removable storage devices, ROM, RAM, magnetic disks, or optical disks and other media that can store program codes.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Multimedia (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Image Analysis (AREA)

Abstract

La présente invention concerne une technique d'intelligence artificielle, fournissant un procédé, un système et un support de stockage de reconnaissance de cible d'image basée sur YOLO, le procédé comprenant les étapes consistant à : recevoir une image à détecter (S102) ; sur la base d'une exigence prédéfinie, régler la taille de l'image à détecter pour générer une première image de détection (S104) ; envoyer la première image de détection à un modèle de réseau neuronal pour mettre en œuvre une reconnaissance de correspondance, et générer une trame de détection et des informations de reconnaissance de classe, et une valeur de probabilité de classe correspondant aux informations de reconnaissance de classe (S106) ; déterminer si la valeur de probabilité de classe est supérieure à une valeur de probabilité de classe prédéfinie (S108) ; si tel est le cas, définir ensuite la trame de détection et les informations de reconnaissance de classe comme résultat de classe reconnu (S110). Le présent procédé permet d'améliorer efficacement la précision de détection et de réduire le délai de détection.
PCT/CN2019/118499 2019-02-14 2019-11-14 Procédé et appareil de reconnaissance de cible d'image basée sur yolo, dispositif électronique et support de stockage WO2020164282A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910114621.5 2019-02-14
CN201910114621.5A CN109977943B (zh) 2019-02-14 2019-02-14 一种基于yolo的图像目标识别方法、系统和存储介质

Publications (1)

Publication Number Publication Date
WO2020164282A1 true WO2020164282A1 (fr) 2020-08-20

Family

ID=67076997

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/118499 WO2020164282A1 (fr) 2019-02-14 2019-11-14 Procédé et appareil de reconnaissance de cible d'image basée sur yolo, dispositif électronique et support de stockage

Country Status (2)

Country Link
CN (1) CN109977943B (fr)
WO (1) WO2020164282A1 (fr)

Cited By (108)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111986255A (zh) * 2020-09-07 2020-11-24 北京凌云光技术集团有限责任公司 一种图像检测模型的多尺度anchor初始化方法与装置
CN112036507A (zh) * 2020-09-25 2020-12-04 北京小米松果电子有限公司 图像识别模型的训练方法、装置、存储介质和电子设备
CN112036286A (zh) * 2020-08-25 2020-12-04 北京华正明天信息技术股份有限公司 一种基于yoloV3算法实现温度感应及智能分析识别火焰的方法
CN112101134A (zh) * 2020-08-24 2020-12-18 深圳市商汤科技有限公司 物体的检测方法及装置、电子设备和存储介质
CN112149748A (zh) * 2020-09-28 2020-12-29 商汤集团有限公司 图像分类方法及装置、电子设备和存储介质
CN112183358A (zh) * 2020-09-29 2021-01-05 新石器慧拓(北京)科技有限公司 一种目标检测模型的训练方法及装置
CN112200186A (zh) * 2020-10-15 2021-01-08 上海海事大学 基于改进yolo_v3模型的车标识别方法
CN112231497A (zh) * 2020-10-19 2021-01-15 腾讯科技（深圳）有限公司 信息分类方法、装置、存储介质及电子设备
CN112288003A (zh) * 2020-10-28 2021-01-29 北京奇艺世纪科技有限公司 一种神经网络训练、及目标检测方法和装置
CN112287884A (zh) * 2020-11-19 2021-01-29 长江大学 一种考试异常行为检测方法、装置及计算机可读存储介质
CN112348778A (zh) * 2020-10-21 2021-02-09 深圳市优必选科技股份有限公司 一种物体识别方法、装置、终端设备及存储介质
CN112348112A (zh) * 2020-11-24 2021-02-09 深圳市优必选科技股份有限公司 图像识别模型的训练方法、训练装置及终端设备
CN112364807A (zh) * 2020-11-24 2021-02-12 深圳市优必选科技股份有限公司 图像识别方法、装置、终端设备及计算机可读存储介质
CN112365465A (zh) * 2020-11-09 2021-02-12 浙江大华技术股份有限公司 合成图像类别确定方法、装置、存储介质及电子装置
CN112507912A (zh) * 2020-12-15 2021-03-16 网易（杭州）网络有限公司 一种识别违规图片的方法及装置
CN112529020A (zh) * 2020-12-24 2021-03-19 携程旅游信息技术（上海）有限公司 基于神经网络的动物识别方法、系统、设备及存储介质
CN112541483A (zh) * 2020-12-25 2021-03-23 三峡大学 Yolo和分块-融合策略结合的稠密人脸检测方法
CN112560586A (zh) * 2020-11-27 2021-03-26 国家电网有限公司大数据中心 一种杆塔标识牌结构化数据获得方法、装置及电子设备
CN112560799A (zh) * 2021-01-05 2021-03-26 北京航空航天大学 基于自适应目标区域搜索和博弈的无人机智能车辆目标检测方法和应用
CN112580523A (zh) * 2020-12-22 2021-03-30 平安国际智慧城市科技股份有限公司 行为识别方法、装置、设备及存储介质
CN112580734A (zh) * 2020-12-25 2021-03-30 深圳市优必选科技股份有限公司 目标检测模型训练方法、系统、终端设备及存储介质
CN112597915A (zh) * 2020-12-26 2021-04-02 上海有个机器人有限公司 对室内近距离行人进行识别的方法、装置、介质和机器人
CN112613097A (zh) * 2020-12-15 2021-04-06 中铁二十四局集团江苏工程有限公司 一种基于计算机视觉的bim快速化建模方法
CN112634327A (zh) * 2020-12-21 2021-04-09 合肥讯图信息科技有限公司 基于YOLOv4模型的跟踪方法
CN112633159A (zh) * 2020-12-22 2021-04-09 北京迈格威科技有限公司 人-物交互关系识别方法、模型训练方法及对应装置
CN112633286A (zh) * 2020-12-25 2021-04-09 北京航星机器制造有限公司 一种基于危险品相似率和识别概率的智能安检系统
CN112634202A (zh) * 2020-12-04 2021-04-09 浙江省农业科学院 一种基于YOLOv3-Lite的混养鱼群行为检测的方法、装置及系统
CN112633352A (zh) * 2020-12-18 2021-04-09 浙江大华技术股份有限公司 一种目标检测方法、装置、电子设备及存储介质
CN112699925A (zh) * 2020-12-23 2021-04-23 国网安徽省电力有限公司检修分公司 一种变电站表计图像分类方法
CN112733741A (zh) * 2021-01-14 2021-04-30 苏州挚途科技有限公司 交通标识牌识别方法、装置和电子设备
CN112734641A (zh) * 2020-12-31 2021-04-30 百果园技术(新加坡)有限公司 目标检测模型的训练方法、装置、计算机设备及介质
CN112766170A (zh) * 2021-01-21 2021-05-07 广西财经学院 基于簇类无人机图像的自适应分割检测方法及装置
CN112784694A (zh) * 2020-12-31 2021-05-11 杭州电子科技大学 一种基于evp_yolo的室内物品检测方法
CN112800971A (zh) * 2021-01-29 2021-05-14 深圳市商汤科技有限公司 神经网络训练及点云数据处理方法、装置、设备和介质
CN112818980A (zh) * 2021-01-15 2021-05-18 湖南千盟物联信息技术有限公司 一种基于Yolov3算法的钢包号检测识别方法
CN112861711A (zh) * 2021-02-05 2021-05-28 深圳市安软科技股份有限公司 区域入侵检测方法、装置、电子设备及存储介质
CN112861716A (zh) * 2021-02-05 2021-05-28 深圳市安软科技股份有限公司 违规物品摆放的监测方法、系统、设备及存储介质
CN112906621A (zh) * 2021-03-10 2021-06-04 北京华捷艾米科技有限公司 一种手部检测方法、装置、存储介质和设备
CN112906478A (zh) * 2021-01-22 2021-06-04 北京百度网讯科技有限公司 目标对象的识别方法、装置、设备和存储介质
CN112906794A (zh) * 2021-02-22 2021-06-04 珠海格力电器股份有限公司 一种目标检测方法、装置、存储介质及终端
CN112911171A (zh) * 2021-02-04 2021-06-04 上海航天控制技术研究所 一种基于加速处理的智能光电信息处理系统及方法
CN112906495A (zh) * 2021-01-27 2021-06-04 深圳安智杰科技有限公司 一种目标检测方法、装置、电子设备及存储介质
CN112966762A (zh) * 2021-03-16 2021-06-15 南京恩博科技有限公司 一种野生动物检测方法、装置、存储介质及电子设备
CN112966618A (zh) * 2021-03-11 2021-06-15 京东数科海益信息科技有限公司 着装识别方法、装置、设备及计算机可读介质
CN112990334A (zh) * 2021-03-29 2021-06-18 西安电子科技大学 基于改进原型网络的小样本sar图像目标识别方法
CN112989924A (zh) * 2021-01-26 2021-06-18 深圳市优必选科技股份有限公司 目标检测方法、目标检测装置及终端设备
CN112991304A (zh) * 2021-03-23 2021-06-18 武汉大学 一种基于激光定向能量沉积监测系统的熔池溅射检测方法
CN113011319A (zh) * 2021-03-16 2021-06-22 上海应用技术大学 多尺度火灾目标识别方法及系统
CN113052127A (zh) * 2021-04-09 2021-06-29 上海云从企业发展有限公司 一种行为检测方法、系统、计算机设备及机器可读介质
CN113095133A (zh) * 2021-03-04 2021-07-09 北京迈格威科技有限公司 模型训练方法、目标检测方法及对应装置
CN113128522A (zh) * 2021-05-11 2021-07-16 四川云从天府人工智能科技有限公司 目标识别方法、装置、计算机设备和存储介质
CN113139597A (zh) * 2021-04-19 2021-07-20 中国人民解放军91054部队 一种基于统计思想的图像分布外检测方法
CN113205067A (zh) * 2021-05-26 2021-08-03 北京京东乾石科技有限公司 作业人员监控方法、装置、电子设备和存储介质
CN113222889A (zh) * 2021-03-30 2021-08-06 大连智慧渔业科技有限公司 高分辨率图像下水产养殖物的工厂化养殖计数方法及装置
CN113240638A (zh) * 2021-05-12 2021-08-10 上海联影智能医疗科技有限公司 基于深度学习的目标检测方法、设备及介质
CN113377888A (zh) * 2021-06-25 2021-09-10 北京百度网讯科技有限公司 训练目标检测模型和检测目标的方法
CN113392833A (zh) * 2021-06-10 2021-09-14 沈阳派得林科技有限责任公司 一种工业射线底片图像铅字编号识别方法
CN113435260A (zh) * 2021-06-07 2021-09-24 上海商汤智能科技有限公司 图像检测方法和相关训练方法及相关装置、设备及介质
CN113486746A (zh) * 2021-06-25 2021-10-08 海南电网有限责任公司三亚供电局 基于生物感应及视频监控的电力电缆防外破方法
CN113486857A (zh) * 2021-08-03 2021-10-08 云南大学 一种基于YOLOv4的登高安全检测方法及系统
CN113536963A (zh) * 2021-06-25 2021-10-22 西安电子科技大学 基于轻量化yolo网络的sar图像飞机目标检测方法
CN113553948A (zh) * 2021-07-23 2021-10-26 中远海运科技(北京)有限公司 烟虫自动识别和计数方法、计算机可读介质
CN113591566A (zh) * 2021-06-28 2021-11-02 北京百度网讯科技有限公司 图像识别模型的训练方法、装置、电子设备和存储介质
CN113723217A (zh) * 2021-08-09 2021-11-30 南京邮电大学 一种基于yolo改进的物体智能检测方法及系统
CN113723157A (zh) * 2020-12-15 2021-11-30 京东数字科技控股股份有限公司 一种农作物病害识别方法、装置、电子设备及存储介质
CN113723406A (zh) * 2021-09-03 2021-11-30 乐普（北京）医疗器械股份有限公司 一种对冠脉造影图像进行支架定位的处理方法和装置
CN113743339A (zh) * 2021-09-09 2021-12-03 三峡大学 一种基于场景识别的室内跌倒检测方法和系统
CN113762023A (zh) * 2021-02-18 2021-12-07 北京京东振世信息技术有限公司 基于物品关联关系的对象识别的方法和装置
CN113792656A (zh) * 2021-09-15 2021-12-14 山东大学 一种人员移动中使用移动通讯设备的行为检测及报警系统
CN113948190A (zh) * 2021-09-02 2022-01-18 上海健康医学院 X线头颅正位片头影测量标志点自动识别方法及设备
CN113989939A (zh) * 2021-11-16 2022-01-28 河北工业大学 一种基于改进yolo算法的小目标行人检测系统
CN114022705A (zh) * 2021-10-29 2022-02-08 电子科技大学 一种基于场景复杂度预分类的自适应目标检测方法
CN114022554A (zh) * 2021-11-03 2022-02-08 北华航天工业学院 一种基于yolo的按摩机器人穴位检测与定位方法
CN114120358A (zh) * 2021-11-11 2022-03-01 国网江苏省电力有限公司技能培训中心 一种基于超像素引导深度学习的人员头戴安全帽识别方法
CN114119455A (zh) * 2021-09-03 2022-03-01 乐普（北京）医疗器械股份有限公司 一种基于目标检测网络定位血管狭窄部位的方法和装置
CN114255389A (zh) * 2021-11-15 2022-03-29 浙江时空道宇科技有限公司 一种目标对象检测方法、装置、设备和存储介质
CN114373075A (zh) * 2021-12-31 2022-04-19 西安电子科技大学广州研究院 目标部件检测数据集的构建方法、检测方法、装置及设备
US20220130139A1 (en) * 2022-01-05 2022-04-28 Baidu Usa Llc Image processing method and apparatus, electronic device and storage medium
CN114565848A (zh) * 2022-02-25 2022-05-31 佛山读图科技有限公司 一种复杂场景的药液液位检测方法及系统
CN114662594A (zh) * 2022-03-25 2022-06-24 浙江省通信产业服务有限公司 一种目标特征识别分析系统
CN114742204A (zh) * 2022-04-08 2022-07-12 黑龙江惠达科技发展有限公司 检测秸秆覆盖率的方法和装置
CN114782778A (zh) * 2022-04-25 2022-07-22 广东工业大学 一种基于机器视觉技术的装配状态监控方法及系统
CN114821288A (zh) * 2021-01-29 2022-07-29 中强光电股份有限公司 图像辨识方法以及无人机系统
CN114842315A (zh) * 2022-05-07 2022-08-02 无锡雪浪数制科技有限公司 轻量化高铁轮毂垫片防松识别方法及装置
CN114881763A (zh) * 2022-05-18 2022-08-09 中国工商银行股份有限公司 养殖业贷后监管方法、装置、设备和介质
CN114972891A (zh) * 2022-07-07 2022-08-30 智云数创(洛阳)数字科技有限公司 一种cad构件自动识别方法及bim建模方法
CN115029209A (zh) * 2022-06-17 2022-09-09 杭州天杭空气质量检测有限公司 一种菌落图像采集处理装置及其处理方法
CN115082661A (zh) * 2022-07-11 2022-09-20 阿斯曼尔科技(上海)有限公司 一种传感器装配难度降低方法
CN115187982A (zh) * 2022-07-12 2022-10-14 河北华清环境科技集团股份有限公司 藻类检测方法、装置及终端设备
CN115297263A (zh) * 2022-08-24 2022-11-04 广州方图科技有限公司 适用于拍立方的自动拍照控制方法、系统及拍立方
CN115346170A (zh) * 2022-08-11 2022-11-15 北京市燃气集团有限责任公司 一种燃气设施区域的智能监控方法及装置
CN115346172A (zh) * 2022-08-16 2022-11-15 哈尔滨市科佳通用机电股份有限公司 一种钩提杆复位弹簧丢失和折断检测方法及系统
WO2022252089A1 (fr) * 2021-05-31 2022-12-08 京东方科技集团股份有限公司 Procédé de formation pour un modèle de détection d'objet, et procédé et dispositif de détection d'objet
CN115546566A (zh) * 2022-11-24 2022-12-30 杭州心识宇宙科技有限公司 基于物品识别的智能体交互方法、装置、设备及存储介质
CN115690565A (zh) * 2022-09-28 2023-02-03 大连海洋大学 融合知识与改进YOLOv5的养殖红鳍东方鲀目标检测方法
CN115690570A (zh) * 2023-01-05 2023-02-03 中国水产科学研究院黄海水产研究所 一种基于st-gcn的鱼群摄食强度预测方法
CN115909358A (zh) * 2022-07-27 2023-04-04 广州市玄武无线科技股份有限公司 商品规格识别方法、装置、终端设备及计算机存储介质
CN116051985A (zh) * 2022-12-20 2023-05-02 中国科学院空天信息创新研究院 一种基于多模型互馈学习的半监督遥感目标检测方法
CN116403163A (zh) * 2023-04-20 2023-07-07 慧铁科技有限公司 一种截断塞门手把开合状态的识别方法和装置
CN116452858A (zh) * 2023-03-24 2023-07-18 哈尔滨市科佳通用机电股份有限公司 一种铁路货车连接拉杆圆销折断故障识别方法及系统
CN116681687A (zh) * 2023-06-20 2023-09-01 广东电网有限责任公司广州供电局 基于计算机视觉的导线检测方法、装置和计算机设备
CN116758547A (zh) * 2023-06-27 2023-09-15 北京中超伟业信息安全技术股份有限公司 一种纸介质碳化方法、系统及存储介质
CN116916166A (zh) * 2023-09-12 2023-10-20 湖南湘银河传感科技有限公司 一种基于ai图像解析的遥测终端机
CN116935232A (zh) * 2023-09-15 2023-10-24 青岛国测海遥信息技术有限公司 海上风力设备的遥感图像处理方法和装置、设备和介质
CN117523318A (zh) * 2023-12-26 2024-02-06 宁波微科光电股份有限公司 一种抗光干扰的地铁屏蔽门异物检测方法、装置及介质
CN117671597A (zh) * 2023-12-25 2024-03-08 北京大学长沙计算与数字经济研究院 一种老鼠检测模型的构建方法和老鼠检测方法及装置
CN117893895A (zh) * 2024-03-15 2024-04-16 山东省海洋资源与环境研究院(山东省海洋环境监测中心、山东省水产品质量检验中心) 一种三疣梭子蟹的识别方法、系统、设备和存储介质
CN112529020B (zh) * 2020-12-24 2024-05-24 携程旅游信息技术（上海）有限公司 基于神经网络的动物识别方法、系统、设备及存储介质

Families Citing this family (27)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109977943B (zh) * 2019-02-14 2024-05-07 平安科技（深圳）有限公司 一种基于yolo的图像目标识别方法、系统和存储介质
CN110348304A (zh) * 2019-06-06 2019-10-18 武汉理工大学 一种可搭载于无人机的海事遇险人员搜索系统以及目标识别方法
CN110738125B (zh) * 2019-09-19 2023-08-01 平安科技（深圳）有限公司 利用Mask R-CNN选择检测框的方法、装置及存储介质
CN111223343B (zh) * 2020-03-07 2022-01-28 上海中科教育装备集团有限公司 一种杠杆平衡实验人工智能评分实验器材及评分方法
CN111582021A (zh) * 2020-03-26 2020-08-25 平安科技（深圳）有限公司 场景图像中的文本检测方法、装置及计算机设备
CN111695559B (zh) * 2020-04-28 2023-07-18 深圳市跨越新科技有限公司 基于YoloV3模型的运单图片信息打码方法及系统
CN113705591A (zh) * 2020-05-20 2021-11-26 上海微创卜算子医疗科技有限公司 可读存储介质、支架规格识别方法及装置
CN111626256B (zh) * 2020-06-03 2023-06-27 兰波(苏州)智能科技有限公司 基于扫描电镜图像的高精度硅藻检测识别方法及系统
CN111738259A (zh) * 2020-06-29 2020-10-02 广东电网有限责任公司 一种杆塔状态检测方法及装置
CN111523621B (zh) * 2020-07-03 2020-10-20 腾讯科技（深圳）有限公司 图像识别方法、装置、计算机设备和存储介质
CN111857350A (zh) * 2020-07-28 2020-10-30 海尔优家智能科技（北京）有限公司 用于旋转显示设备的方法及装置、设备
CN112132018A (zh) * 2020-09-22 2020-12-25 平安国际智慧城市科技股份有限公司 交警识别方法、装置、介质及电子设备
CN112116582A (zh) * 2020-09-24 2020-12-22 深圳爱莫科技有限公司 一种库存或陈列场景下的条烟检测识别方法
CN112132088B (zh) * 2020-09-29 2024-01-12 动联(山东)电子科技有限公司 一种巡检点位漏巡识别方法
CN112381773B (zh) * 2020-11-05 2023-04-18 东风柳州汽车有限公司 关键截面数据分析方法、装置、设备及存储介质
CN112508915A (zh) * 2020-12-11 2021-03-16 中信银行股份有限公司 一种目标检测结果优化方法及系统
CN112215308B (zh) * 2020-12-13 2021-03-30 之江实验室 一种吊装物体单阶检测方法、装置、电子设备及存储介质
CN112507896B (zh) * 2020-12-14 2023-11-07 大连大学 一种采用改进的yolo-v4模型对樱桃果实进行检测的方法
CN112613570A (zh) * 2020-12-29 2021-04-06 深圳云天励飞技术股份有限公司 一种图像检测方法、图像检测装置、设备及存储介质
CN113033398B (zh) * 2021-03-25 2022-02-11 深圳市康冠商用科技有限公司 一种手势识别方法、装置、计算机设备及存储介质
CN112965604A (zh) * 2021-03-29 2021-06-15 深圳市优必选科技股份有限公司 手势识别方法、装置、终端设备及计算机可读存储介质
CN113158922A (zh) * 2021-04-26 2021-07-23 平安科技（深圳）有限公司 基于yolo神经网络的车流量统计方法、装置及设备
CN113269188B (zh) * 2021-06-17 2023-03-14 华南农业大学 一种标记点及其像素坐标检测方法
CN113705643B (zh) * 2021-08-17 2022-10-28 荣耀终端有限公司 一种目标物检测方法、装置以及电子设备
CN113657280A (zh) * 2021-08-18 2021-11-16 广东电网有限责任公司 一种输电线路目标缺陷检测示警方法及系统
CN116342316A (zh) * 2023-05-31 2023-06-27 青岛希尔信息科技有限公司 会计核算和项目财务管理系统及方法
CN117201834A (zh) * 2023-09-11 2023-12-08 南京天创电子技术有限公司 基于目标检测的实时双光谱融合视频流显示方法及系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107527009A (zh) * 2017-07-11 2017-12-29 浙江汉凡软件科技有限公司 一种基于yolo目标检测的遗留物检测方法
CN108154098A (zh) * 2017-12-20 2018-06-12 歌尔股份有限公司 一种机器人的目标识别方法、装置和机器人
CN109117794A (zh) * 2018-08-16 2019-01-01 广东工业大学 一种运动目标行为跟踪方法、装置、设备及可读存储介质
CN109977943A (zh) * 2019-02-14 2019-07-05 平安科技（深圳）有限公司 一种基于yolo的图像目标识别方法、系统和存储介质

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107247956B (zh) * 2016-10-09 2020-03-27 成都快眼科技有限公司 一种基于网格判断的快速目标检测方法
CN107423760A (zh) * 2017-07-21 2017-12-01 西安电子科技大学 基于预分割和回归的深度学习目标检测方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107527009A (zh) * 2017-07-11 2017-12-29 浙江汉凡软件科技有限公司 一种基于yolo目标检测的遗留物检测方法
CN108154098A (zh) * 2017-12-20 2018-06-12 歌尔股份有限公司 一种机器人的目标识别方法、装置和机器人
CN109117794A (zh) * 2018-08-16 2019-01-01 广东工业大学 一种运动目标行为跟踪方法、装置、设备及可读存储介质
CN109977943A (zh) * 2019-02-14 2019-07-05 平安科技（深圳）有限公司 一种基于yolo的图像目标识别方法、系统和存储介质

Cited By (175)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112101134B (zh) * 2020-08-24 2024-01-02 深圳市商汤科技有限公司 物体的检测方法及装置、电子设备和存储介质
CN112101134A (zh) * 2020-08-24 2020-12-18 深圳市商汤科技有限公司 物体的检测方法及装置、电子设备和存储介质
CN112036286A (zh) * 2020-08-25 2020-12-04 北京华正明天信息技术股份有限公司 一种基于yoloV3算法实现温度感应及智能分析识别火焰的方法
CN111986255B (zh) * 2020-09-07 2024-04-09 凌云光技术股份有限公司 一种图像检测模型的多尺度anchor初始化方法与装置
CN111986255A (zh) * 2020-09-07 2020-11-24 北京凌云光技术集团有限责任公司 一种图像检测模型的多尺度anchor初始化方法与装置
CN112036507B (zh) * 2020-09-25 2023-11-14 北京小米松果电子有限公司 图像识别模型的训练方法、装置、存储介质和电子设备
CN112036507A (zh) * 2020-09-25 2020-12-04 北京小米松果电子有限公司 图像识别模型的训练方法、装置、存储介质和电子设备
CN112149748A (zh) * 2020-09-28 2020-12-29 商汤集团有限公司 图像分类方法及装置、电子设备和存储介质
CN112149748B (zh) * 2020-09-28 2024-05-21 商汤集团有限公司 图像分类方法及装置、电子设备和存储介质
CN112183358A (zh) * 2020-09-29 2021-01-05 新石器慧拓(北京)科技有限公司 一种目标检测模型的训练方法及装置
CN112183358B (zh) * 2020-09-29 2024-04-23 新石器慧通（北京）科技有限公司 一种目标检测模型的训练方法及装置
CN112200186A (zh) * 2020-10-15 2021-01-08 上海海事大学 基于改进yolo_v3模型的车标识别方法
CN112200186B (zh) * 2020-10-15 2024-03-15 上海海事大学 基于改进yolo_v3模型的车标识别方法
CN112231497B (zh) * 2020-10-19 2024-04-09 腾讯科技（深圳）有限公司 信息分类方法、装置、存储介质及电子设备
CN112231497A (zh) * 2020-10-19 2021-01-15 腾讯科技（深圳）有限公司 信息分类方法、装置、存储介质及电子设备
CN112348778A (zh) * 2020-10-21 2021-02-09 深圳市优必选科技股份有限公司 一种物体识别方法、装置、终端设备及存储介质
CN112348778B (zh) * 2020-10-21 2023-10-27 深圳市优必选科技股份有限公司 一种物体识别方法、装置、终端设备及存储介质
CN112288003A (zh) * 2020-10-28 2021-01-29 北京奇艺世纪科技有限公司 一种神经网络训练、及目标检测方法和装置
CN112365465A (zh) * 2020-11-09 2021-02-12 浙江大华技术股份有限公司 合成图像类别确定方法、装置、存储介质及电子装置
CN112365465B (zh) * 2020-11-09 2024-02-06 浙江大华技术股份有限公司 合成图像类别确定方法、装置、存储介质及电子装置
CN112287884B (zh) * 2020-11-19 2024-02-20 长江大学 一种考试异常行为检测方法、装置及计算机可读存储介质
CN112287884A (zh) * 2020-11-19 2021-01-29 长江大学 一种考试异常行为检测方法、装置及计算机可读存储介质
CN112348112B (zh) * 2020-11-24 2023-12-15 深圳市优必选科技股份有限公司 图像识别模型的训练方法、训练装置及终端设备
CN112364807B (zh) * 2020-11-24 2023-12-15 深圳市优必选科技股份有限公司 图像识别方法、装置、终端设备及计算机可读存储介质
CN112364807A (zh) * 2020-11-24 2021-02-12 深圳市优必选科技股份有限公司 图像识别方法、装置、终端设备及计算机可读存储介质
CN112348112A (zh) * 2020-11-24 2021-02-09 深圳市优必选科技股份有限公司 图像识别模型的训练方法、训练装置及终端设备
CN112560586B (zh) * 2020-11-27 2024-05-10 国家电网有限公司大数据中心 一种杆塔标识牌结构化数据获得方法、装置及电子设备
CN112560586A (zh) * 2020-11-27 2021-03-26 国家电网有限公司大数据中心 一种杆塔标识牌结构化数据获得方法、装置及电子设备
CN112634202A (zh) * 2020-12-04 2021-04-09 浙江省农业科学院 一种基于YOLOv3-Lite的混养鱼群行为检测的方法、装置及系统
CN112613097A (zh) * 2020-12-15 2021-04-06 中铁二十四局集团江苏工程有限公司 一种基于计算机视觉的bim快速化建模方法
CN113723157B (zh) * 2020-12-15 2024-02-09 京东科技控股股份有限公司 一种农作物病害识别方法、装置、电子设备及存储介质
CN113723157A (zh) * 2020-12-15 2021-11-30 京东数字科技控股股份有限公司 一种农作物病害识别方法、装置、电子设备及存储介质
CN112507912A (zh) * 2020-12-15 2021-03-16 网易（杭州）网络有限公司 一种识别违规图片的方法及装置
CN112633352B (zh) * 2020-12-18 2023-08-29 浙江大华技术股份有限公司 一种目标检测方法、装置、电子设备及存储介质
CN112633352A (zh) * 2020-12-18 2021-04-09 浙江大华技术股份有限公司 一种目标检测方法、装置、电子设备及存储介质
CN112634327A (zh) * 2020-12-21 2021-04-09 合肥讯图信息科技有限公司 基于YOLOv4模型的跟踪方法
CN112633159B (zh) * 2020-12-22 2024-04-12 北京迈格威科技有限公司 人-物交互关系识别方法、模型训练方法及对应装置
CN112633159A (zh) * 2020-12-22 2021-04-09 北京迈格威科技有限公司 人-物交互关系识别方法、模型训练方法及对应装置
CN112580523A (zh) * 2020-12-22 2021-03-30 平安国际智慧城市科技股份有限公司 行为识别方法、装置、设备及存储介质
CN112699925A (zh) * 2020-12-23 2021-04-23 国网安徽省电力有限公司检修分公司 一种变电站表计图像分类方法
CN112529020A (zh) * 2020-12-24 2021-03-19 携程旅游信息技术（上海）有限公司 基于神经网络的动物识别方法、系统、设备及存储介质
CN112529020B (zh) * 2020-12-24 2024-05-24 携程旅游信息技术（上海）有限公司 基于神经网络的动物识别方法、系统、设备及存储介质
CN112580734A (zh) * 2020-12-25 2021-03-30 深圳市优必选科技股份有限公司 目标检测模型训练方法、系统、终端设备及存储介质
CN112580734B (zh) * 2020-12-25 2023-12-29 深圳市优必选科技股份有限公司 目标检测模型训练方法、系统、终端设备及存储介质
CN112541483A (zh) * 2020-12-25 2021-03-23 三峡大学 Yolo和分块-融合策略结合的稠密人脸检测方法
CN112633286A (zh) * 2020-12-25 2021-04-09 北京航星机器制造有限公司 一种基于危险品相似率和识别概率的智能安检系统
CN112541483B (zh) * 2020-12-25 2024-05-17 深圳市富浩鹏电子有限公司 Yolo和分块-融合策略结合的稠密人脸检测方法
CN112597915B (zh) * 2020-12-26 2024-04-09 上海有个机器人有限公司 对室内近距离行人进行识别的方法、装置、介质和机器人
CN112597915A (zh) * 2020-12-26 2021-04-02 上海有个机器人有限公司 对室内近距离行人进行识别的方法、装置、介质和机器人
CN112784694A (zh) * 2020-12-31 2021-05-11 杭州电子科技大学 一种基于evp_yolo的室内物品检测方法
CN112734641A (zh) * 2020-12-31 2021-04-30 百果园技术(新加坡)有限公司 目标检测模型的训练方法、装置、计算机设备及介质
CN112734641B (zh) * 2020-12-31 2024-05-31 百果园技术(新加坡)有限公司 目标检测模型的训练方法、装置、计算机设备及介质
CN112560799A (zh) * 2021-01-05 2021-03-26 北京航空航天大学 基于自适应目标区域搜索和博弈的无人机智能车辆目标检测方法和应用
CN112560799B (zh) * 2021-01-05 2022-08-05 北京航空航天大学 基于自适应目标区域搜索和博弈的无人机智能车辆目标检测方法和应用
CN112733741A (zh) * 2021-01-14 2021-04-30 苏州挚途科技有限公司 交通标识牌识别方法、装置和电子设备
CN112818980A (zh) * 2021-01-15 2021-05-18 湖南千盟物联信息技术有限公司 一种基于Yolov3算法的钢包号检测识别方法
CN112766170B (zh) * 2021-01-21 2024-04-16 广西财经学院 基于簇类无人机图像的自适应分割检测方法及装置
CN112766170A (zh) * 2021-01-21 2021-05-07 广西财经学院 基于簇类无人机图像的自适应分割检测方法及装置
CN112906478B (zh) * 2021-01-22 2024-01-09 北京百度网讯科技有限公司 目标对象的识别方法、装置、设备和存储介质
CN112906478A (zh) * 2021-01-22 2021-06-04 北京百度网讯科技有限公司 目标对象的识别方法、装置、设备和存储介质
CN112989924B (zh) * 2021-01-26 2024-05-24 深圳市优必选科技股份有限公司 目标检测方法、目标检测装置及终端设备
CN112989924A (zh) * 2021-01-26 2021-06-18 深圳市优必选科技股份有限公司 目标检测方法、目标检测装置及终端设备
CN112906495B (zh) * 2021-01-27 2024-04-30 深圳安智杰科技有限公司 一种目标检测方法、装置、电子设备及存储介质
CN112906495A (zh) * 2021-01-27 2021-06-04 深圳安智杰科技有限公司 一种目标检测方法、装置、电子设备及存储介质
CN112800971A (zh) * 2021-01-29 2021-05-14 深圳市商汤科技有限公司 神经网络训练及点云数据处理方法、装置、设备和介质
CN114821288A (zh) * 2021-01-29 2022-07-29 中强光电股份有限公司 图像辨识方法以及无人机系统
CN112911171B (zh) * 2021-02-04 2022-04-22 上海航天控制技术研究所 一种基于加速处理的智能光电信息处理系统及方法
CN112911171A (zh) * 2021-02-04 2021-06-04 上海航天控制技术研究所 一种基于加速处理的智能光电信息处理系统及方法
CN112861711A (zh) * 2021-02-05 2021-05-28 深圳市安软科技股份有限公司 区域入侵检测方法、装置、电子设备及存储介质
CN112861716A (zh) * 2021-02-05 2021-05-28 深圳市安软科技股份有限公司 违规物品摆放的监测方法、系统、设备及存储介质
CN113762023A (zh) * 2021-02-18 2021-12-07 北京京东振世信息技术有限公司 基于物品关联关系的对象识别的方法和装置
CN113762023B (zh) * 2021-02-18 2024-05-24 北京京东振世信息技术有限公司 基于物品关联关系的对象识别的方法和装置
CN112906794A (zh) * 2021-02-22 2021-06-04 珠海格力电器股份有限公司 一种目标检测方法、装置、存储介质及终端
CN113095133B (zh) * 2021-03-04 2023-12-29 北京迈格威科技有限公司 模型训练方法、目标检测方法及对应装置
CN113095133A (zh) * 2021-03-04 2021-07-09 北京迈格威科技有限公司 模型训练方法、目标检测方法及对应装置
CN112906621A (zh) * 2021-03-10 2021-06-04 北京华捷艾米科技有限公司 一种手部检测方法、装置、存储介质和设备
CN112966618A (zh) * 2021-03-11 2021-06-15 京东数科海益信息科技有限公司 着装识别方法、装置、设备及计算机可读介质
CN112966618B (zh) * 2021-03-11 2024-02-09 京东科技信息技术有限公司 着装识别方法、装置、设备及计算机可读介质
CN112966762B (zh) * 2021-03-16 2023-12-26 南京恩博科技有限公司 一种野生动物检测方法、装置、存储介质及电子设备
CN113011319A (zh) * 2021-03-16 2021-06-22 上海应用技术大学 多尺度火灾目标识别方法及系统
CN112966762A (zh) * 2021-03-16 2021-06-15 南京恩博科技有限公司 一种野生动物检测方法、装置、存储介质及电子设备
CN113011319B (zh) * 2021-03-16 2024-04-16 上海应用技术大学 多尺度火灾目标识别方法及系统
CN112991304A (zh) * 2021-03-23 2021-06-18 武汉大学 一种基于激光定向能量沉积监测系统的熔池溅射检测方法
CN112990334A (zh) * 2021-03-29 2021-06-18 西安电子科技大学 基于改进原型网络的小样本sar图像目标识别方法
CN113222889A (zh) * 2021-03-30 2021-08-06 大连智慧渔业科技有限公司 高分辨率图像下水产养殖物的工厂化养殖计数方法及装置
CN113222889B (zh) * 2021-03-30 2024-03-12 大连智慧渔业科技有限公司 高分辨率图像下水产养殖物的工厂化养殖计数方法及装置
CN113052127A (zh) * 2021-04-09 2021-06-29 上海云从企业发展有限公司 一种行为检测方法、系统、计算机设备及机器可读介质
CN113139597A (zh) * 2021-04-19 2021-07-20 中国人民解放军91054部队 一种基于统计思想的图像分布外检测方法
CN113139597B (zh) * 2021-04-19 2022-11-04 中国人民解放军91054部队 一种基于统计思想的图像分布外检测方法
CN113128522B (zh) * 2021-05-11 2024-04-05 四川云从天府人工智能科技有限公司 目标识别方法、装置、计算机设备和存储介质
CN113128522A (zh) * 2021-05-11 2021-07-16 四川云从天府人工智能科技有限公司 目标识别方法、装置、计算机设备和存储介质
CN113240638A (zh) * 2021-05-12 2021-08-10 上海联影智能医疗科技有限公司 基于深度学习的目标检测方法、设备及介质
CN113240638B (zh) * 2021-05-12 2023-11-10 上海联影智能医疗科技有限公司 基于深度学习的目标检测方法、设备及介质
CN113205067B (zh) * 2021-05-26 2024-04-09 北京京东乾石科技有限公司 作业人员监控方法、装置、电子设备和存储介质
CN113205067A (zh) * 2021-05-26 2021-08-03 北京京东乾石科技有限公司 作业人员监控方法、装置、电子设备和存储介质
WO2022252089A1 (fr) * 2021-05-31 2022-12-08 京东方科技集团股份有限公司 Procédé de formation pour un modèle de détection d'objet, et procédé et dispositif de détection d'objet
CN113435260A (zh) * 2021-06-07 2021-09-24 上海商汤智能科技有限公司 图像检测方法和相关训练方法及相关装置、设备及介质
CN113392833A (zh) * 2021-06-10 2021-09-14 沈阳派得林科技有限责任公司 一种工业射线底片图像铅字编号识别方法
CN113377888A (zh) * 2021-06-25 2021-09-10 北京百度网讯科技有限公司 训练目标检测模型和检测目标的方法
CN113536963A (zh) * 2021-06-25 2021-10-22 西安电子科技大学 基于轻量化yolo网络的sar图像飞机目标检测方法
CN113536963B (zh) * 2021-06-25 2023-08-15 西安电子科技大学 基于轻量化yolo网络的sar图像飞机目标检测方法
CN113377888B (zh) * 2021-06-25 2024-04-02 北京百度网讯科技有限公司 训练目标检测模型和检测目标的方法
CN113486746A (zh) * 2021-06-25 2021-10-08 海南电网有限责任公司三亚供电局 基于生物感应及视频监控的电力电缆防外破方法
CN113591566A (zh) * 2021-06-28 2021-11-02 北京百度网讯科技有限公司 图像识别模型的训练方法、装置、电子设备和存储介质
CN113553948A (zh) * 2021-07-23 2021-10-26 中远海运科技(北京)有限公司 烟虫自动识别和计数方法、计算机可读介质
CN113486857A (zh) * 2021-08-03 2021-10-08 云南大学 一种基于YOLOv4的登高安全检测方法及系统
CN113486857B (zh) * 2021-08-03 2023-05-12 云南大学 一种基于YOLOv4的登高安全检测方法及系统
CN113723217A (zh) * 2021-08-09 2021-11-30 南京邮电大学 一种基于yolo改进的物体智能检测方法及系统
CN113948190A (zh) * 2021-09-02 2022-01-18 上海健康医学院 X线头颅正位片头影测量标志点自动识别方法及设备
CN114119455B (zh) * 2021-09-03 2024-04-09 乐普（北京）医疗器械股份有限公司 一种基于目标检测网络定位血管狭窄部位的方法和装置
CN114119455A (zh) * 2021-09-03 2022-03-01 乐普（北京）医疗器械股份有限公司 一种基于目标检测网络定位血管狭窄部位的方法和装置
CN113723406B (zh) * 2021-09-03 2023-07-18 乐普（北京）医疗器械股份有限公司 一种对冠脉造影图像进行支架定位的处理方法和装置
CN113723406A (zh) * 2021-09-03 2021-11-30 乐普（北京）医疗器械股份有限公司 一种对冠脉造影图像进行支架定位的处理方法和装置
CN113743339B (zh) * 2021-09-09 2023-10-03 三峡大学 一种基于场景识别的室内跌倒检测方法和系统
CN113743339A (zh) * 2021-09-09 2021-12-03 三峡大学 一种基于场景识别的室内跌倒检测方法和系统
CN113792656A (zh) * 2021-09-15 2021-12-14 山东大学 一种人员移动中使用移动通讯设备的行为检测及报警系统
CN113792656B (zh) * 2021-09-15 2023-07-18 山东大学 一种人员移动中使用移动通讯设备的行为检测及报警系统
CN114022705B (zh) * 2021-10-29 2023-08-04 电子科技大学 一种基于场景复杂度预分类的自适应目标检测方法
CN114022705A (zh) * 2021-10-29 2022-02-08 电子科技大学 一种基于场景复杂度预分类的自适应目标检测方法
CN114022554B (zh) * 2021-11-03 2023-02-03 北华航天工业学院 一种基于yolo的按摩机器人穴位检测与定位方法
CN114022554A (zh) * 2021-11-03 2022-02-08 北华航天工业学院 一种基于yolo的按摩机器人穴位检测与定位方法
CN114120358B (zh) * 2021-11-11 2024-04-26 国网江苏省电力有限公司技能培训中心 一种基于超像素引导深度学习的人员头戴安全帽识别方法
CN114120358A (zh) * 2021-11-11 2022-03-01 国网江苏省电力有限公司技能培训中心 一种基于超像素引导深度学习的人员头戴安全帽识别方法
CN114255389A (zh) * 2021-11-15 2022-03-29 浙江时空道宇科技有限公司 一种目标对象检测方法、装置、设备和存储介质
CN113989939A (zh) * 2021-11-16 2022-01-28 河北工业大学 一种基于改进yolo算法的小目标行人检测系统
CN113989939B (zh) * 2021-11-16 2024-05-14 河北工业大学 一种基于改进yolo算法的小目标行人检测系统
CN114373075A (zh) * 2021-12-31 2022-04-19 西安电子科技大学广州研究院 目标部件检测数据集的构建方法、检测方法、装置及设备
US20220130139A1 (en) * 2022-01-05 2022-04-28 Baidu Usa Llc Image processing method and apparatus, electronic device and storage medium
US11756288B2 (en) * 2022-01-05 2023-09-12 Baidu Usa Llc Image processing method and apparatus, electronic device and storage medium
CN114565848A (zh) * 2022-02-25 2022-05-31 佛山读图科技有限公司 一种复杂场景的药液液位检测方法及系统
CN114565848B (zh) * 2022-02-25 2022-12-02 佛山读图科技有限公司 一种复杂场景的药液液位检测方法及系统
CN114662594A (zh) * 2022-03-25 2022-06-24 浙江省通信产业服务有限公司 一种目标特征识别分析系统
CN114662594B (zh) * 2022-03-25 2022-10-04 浙江省通信产业服务有限公司 一种目标特征识别分析系统
CN114742204A (zh) * 2022-04-08 2022-07-12 黑龙江惠达科技发展有限公司 检测秸秆覆盖率的方法和装置
CN114782778A (zh) * 2022-04-25 2022-07-22 广东工业大学 一种基于机器视觉技术的装配状态监控方法及系统
CN114782778B (zh) * 2022-04-25 2023-01-06 广东工业大学 一种基于机器视觉技术的装配状态监控方法及系统
CN114842315A (zh) * 2022-05-07 2022-08-02 无锡雪浪数制科技有限公司 轻量化高铁轮毂垫片防松识别方法及装置
CN114842315B (zh) * 2022-05-07 2024-02-02 无锡雪浪数制科技有限公司 轻量化高铁轮毂垫片防松识别方法及装置
CN114881763A (zh) * 2022-05-18 2022-08-09 中国工商银行股份有限公司 养殖业贷后监管方法、装置、设备和介质
CN114881763B (zh) * 2022-05-18 2023-05-26 中国工商银行股份有限公司 养殖业贷后监管方法、装置、设备和介质
CN115029209A (zh) * 2022-06-17 2022-09-09 杭州天杭空气质量检测有限公司 一种菌落图像采集处理装置及其处理方法
CN114972891A (zh) * 2022-07-07 2022-08-30 智云数创(洛阳)数字科技有限公司 一种cad构件自动识别方法及bim建模方法
CN114972891B (zh) * 2022-07-07 2024-05-03 智云数创(洛阳)数字科技有限公司 一种cad构件自动识别方法及bim建模方法
CN115082661A (zh) * 2022-07-11 2022-09-20 阿斯曼尔科技(上海)有限公司 一种传感器装配难度降低方法
CN115082661B (zh) * 2022-07-11 2024-05-10 阿斯曼尔科技(上海)有限公司 一种传感器装配难度降低方法
CN115187982A (zh) * 2022-07-12 2022-10-14 河北华清环境科技集团股份有限公司 藻类检测方法、装置及终端设备
CN115909358B (zh) * 2022-07-27 2024-02-13 广州市玄武无线科技股份有限公司 商品规格识别方法、装置、终端设备及计算机存储介质
CN115909358A (zh) * 2022-07-27 2023-04-04 广州市玄武无线科技股份有限公司 商品规格识别方法、装置、终端设备及计算机存储介质
CN115346170B (zh) * 2022-08-11 2023-05-30 北京市燃气集团有限责任公司 一种燃气设施区域的智能监控方法及装置
CN115346170A (zh) * 2022-08-11 2022-11-15 北京市燃气集团有限责任公司 一种燃气设施区域的智能监控方法及装置
CN115346172B (zh) * 2022-08-16 2023-04-21 哈尔滨市科佳通用机电股份有限公司 一种钩提杆复位弹簧丢失和折断检测方法及系统
CN115346172A (zh) * 2022-08-16 2022-11-15 哈尔滨市科佳通用机电股份有限公司 一种钩提杆复位弹簧丢失和折断检测方法及系统
CN115297263A (zh) * 2022-08-24 2022-11-04 广州方图科技有限公司 适用于拍立方的自动拍照控制方法、系统及拍立方
CN115297263B (zh) * 2022-08-24 2023-04-07 广州方图科技有限公司 适用于拍立方的自动拍照控制方法、系统及拍立方
CN115690565B (zh) * 2022-09-28 2024-02-20 大连海洋大学 融合知识与改进YOLOv5的养殖红鳍东方鲀目标检测方法
CN115690565A (zh) * 2022-09-28 2023-02-03 大连海洋大学 融合知识与改进YOLOv5的养殖红鳍东方鲀目标检测方法
CN115546566A (zh) * 2022-11-24 2022-12-30 杭州心识宇宙科技有限公司 基于物品识别的智能体交互方法、装置、设备及存储介质
CN116051985B (zh) * 2022-12-20 2023-06-23 中国科学院空天信息创新研究院 一种基于多模型互馈学习的半监督遥感目标检测方法
CN116051985A (zh) * 2022-12-20 2023-05-02 中国科学院空天信息创新研究院 一种基于多模型互馈学习的半监督遥感目标检测方法
CN115690570B (zh) * 2023-01-05 2023-03-28 中国水产科学研究院黄海水产研究所 一种基于st-gcn的鱼群摄食强度预测方法
CN115690570A (zh) * 2023-01-05 2023-02-03 中国水产科学研究院黄海水产研究所 一种基于st-gcn的鱼群摄食强度预测方法
CN116452858B (zh) * 2023-03-24 2023-12-15 哈尔滨市科佳通用机电股份有限公司 一种铁路货车连接拉杆圆销折断故障识别方法及系统
CN116452858A (zh) * 2023-03-24 2023-07-18 哈尔滨市科佳通用机电股份有限公司 一种铁路货车连接拉杆圆销折断故障识别方法及系统
CN116403163B (zh) * 2023-04-20 2023-10-27 慧铁科技有限公司 一种截断塞门手把开合状态的识别方法和装置
CN116403163A (zh) * 2023-04-20 2023-07-07 慧铁科技有限公司 一种截断塞门手把开合状态的识别方法和装置
CN116681687A (zh) * 2023-06-20 2023-09-01 广东电网有限责任公司广州供电局 基于计算机视觉的导线检测方法、装置和计算机设备
CN116758547A (zh) * 2023-06-27 2023-09-15 北京中超伟业信息安全技术股份有限公司 一种纸介质碳化方法、系统及存储介质
CN116758547B (zh) * 2023-06-27 2024-03-12 北京中超伟业信息安全技术股份有限公司 一种纸介质碳化方法、系统及存储介质
CN116916166A (zh) * 2023-09-12 2023-10-20 湖南湘银河传感科技有限公司 一种基于ai图像解析的遥测终端机
CN116916166B (zh) * 2023-09-12 2023-11-17 湖南湘银河传感科技有限公司 一种基于ai图像解析的遥测终端机
CN116935232A (zh) * 2023-09-15 2023-10-24 青岛国测海遥信息技术有限公司 海上风力设备的遥感图像处理方法和装置、设备和介质
CN117671597A (zh) * 2023-12-25 2024-03-08 北京大学长沙计算与数字经济研究院 一种老鼠检测模型的构建方法和老鼠检测方法及装置
CN117523318B (zh) * 2023-12-26 2024-04-16 宁波微科光电股份有限公司 一种抗光干扰的地铁屏蔽门异物检测方法、装置及介质
CN117523318A (zh) * 2023-12-26 2024-02-06 宁波微科光电股份有限公司 一种抗光干扰的地铁屏蔽门异物检测方法、装置及介质
CN117893895A (zh) * 2024-03-15 2024-04-16 山东省海洋资源与环境研究院(山东省海洋环境监测中心、山东省水产品质量检验中心) 一种三疣梭子蟹的识别方法、系统、设备和存储介质

Also Published As

Publication number Publication date
CN109977943A (zh) 2019-07-05
CN109977943B (zh) 2024-05-07

Similar Documents

Publication Publication Date Title
WO2020164282A1 (fr) Procédé et appareil de reconnaissance de cible d'image basée sur yolo, dispositif électronique et support de stockage
CN110533084B (zh) 一种基于自注意力机制的多尺度目标检测方法
CN112052787B (zh) 基于人工智能的目标检测方法、装置及电子设备
CN109978893B (zh) 图像语义分割网络的训练方法、装置、设备及存储介质
CN110941594B (zh) 一种视频文件的拆分方法、装置、电子设备及存储介质
US8401292B2 (en) Identifying high saliency regions in digital images
CN110378297B (zh) 基于深度学习的遥感图像目标检测方法、装置、及存储介质
US20170032247A1 (en) Media classification
CN111079674B (zh) 一种基于全局和局部信息融合的目标检测方法
EP3493101A1 (fr) Procédé de reconnaissance d'images, terminal et support de stockage non volatile
KR20210110823A (ko) 이미지 인식 방법, 인식 모델의 트레이닝 방법 및 관련 장치, 기기
WO2017105655A1 (fr) Procédés pour une localisation d'objet et une classification d'image
CN110991311A (zh) 一种基于密集连接深度网络的目标检测方法
KR20200052439A (ko) 딥러닝 모델의 최적화 시스템 및 방법
CN109871821A (zh) 自适应网络的行人重识别方法、装置、设备及存储介质
WO2021027157A1 (fr) Procédé et appareil d'identification de règlement de déclaration de sinistre de véhicule basé sur l'identification d'image, et dispositif informatique et support de stockage
CN110008899B (zh) 一种可见光遥感图像候选目标提取与分类方法
CN111724342A (zh) 一种用于超声影像中甲状腺微小结节的检测方法
CN112949578B (zh) 车灯状态识别方法、装置、设备及存储介质
CN110069959A (zh) 一种人脸检测方法、装置及用户设备
CN111274972A (zh) 基于度量学习的菜品识别方法及装置
JPWO2015146113A1 (ja) 識別辞書学習システム、識別辞書学習方法および識別辞書学習プログラム
CN114139564B (zh) 二维码检测方法、装置、终端设备及检测网络的训练方法
CN113487610B (zh) 疱疹图像识别方法、装置、计算机设备和存储介质
CN111582057B (zh) 一种基于局部感受野的人脸验证方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19915076

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19915076

Country of ref document: EP

Kind code of ref document: A1