77 results about "Visual information processing" patented technology

The invention discloses a contour and boundary detection algorithm based on a visual color theory and homogeneity inhibition, belongs to the cross field of computer vision and pattern recognition, and aims at extracting the contour and boundary of a target from a complex natural scene. Through the research of the visual information processing mechanism of human eyes, the method carries out the mathematic modeling of a receptive field of a nerve cell of a visual pathway at each stage, and employs the modulation effect of a non-classical receptive field to inhibit the texture edge, thereby highlighting the contour and boundary. The innovation of the invention lies in introducing the visual information processing mechanism of human eyes to a contour and boundary detection model, detecting the color and brightness boundary through setting imbalance vision cone input, maintaining the contour integrity, giving consideration to the homogeneity of the texture region and the homogeneity inhibition on a classical receptive field from the non-classical receptive field, employing the homogeneity inhibition to inhibit the texture edge, and extracting the contour and boundary of a natural image.

The invention discloses a hybrid registration method and system for target objects of a mobile augmented reality (MAR) system. The method comprises the following steps: (1) obtaining current geographical position information and target object information; (2) obtaining the target objects appeared in the current screen of mobile equipment and corresponding screen coordinate values of the target objects according to the current orientation and attitude data of the mobile equipment; (3) obtaining global motion parameters of the current mobile equipment according to the current continuous scene image of the mobile equipment; and (4) adjusting the screen coordinate value in corresponding registration information of each target object appeared in the current screen of the mobile equipment in real time according to the global motion parameters. The system comprises the mobile equipment, and the mobile equipment comprises a sensor unit, a sensor information processing unit, an image acquisition unit, a visual information processing unit and a hybrid registration unit. The hybrid registration method and system have the advantages of good real-time property, fast speed, high accuracy and wide application range.

The invention discloses a fly-simulation visual online detection device and method for surface defects, belonging to the technical field of industrial visual detection and image processing. According to the invention, the surface defects under a complicated background are detected by combining a common CCD (Charge-coupled Device) image sensor with a surface defect detection algorithm. According to the principle of the fly-simulation visual online detection device, the low contrast between the defects and the background needed by the traditional defect detection algorithm is broken through, the obtained scene images can be qualitatively approximate to compound-eye low-resolution images by simulating a fly compound-eye visual information processing method in lower-resolution and low-contrast scene images, and the accuracy is higher. Meanwhile, aiming at the surface defects static in a scene, the fly-simulation visual online detection device also can discover the existence of the surface defects, therefore reliable information is provided for subsequent treatment.

The invention discloses a color constancy method in two steps under complex illumination, belonging to the technical fields of computer vision, image processing and pattern recognition. The invention provides a novel color constancy method in two steps according to principles that a human visual system has independent color passage and bright passage at early stage and visual information processing has modularity. The method comprises the following steps: firstly, utilizing pixels nearby edges to estimate color offset information of an original image, correcting color offset of a color-offset image through white balance; and secondly, according to the color relativity characteristic of three components in an RGB color space, performing same proportion enhancement on the three components of RGB of a color image through obtaining a bright gain curved surface to finally obtain a color image with unchanged color. The method of the invention solves the mal posed problems of traditional color constancy, can preferably maintain the constancy of color under color offset, low luminance and other complex illumination, can effectively improve the contrast and brightness of the image simultaneously and has good universality.

The invention discloses a laser-vision linked night invasion detection device, which realizes the detection and identification of invasion events by laser detection and active vision technology at night. The invention relates to invasion detection technology in security protection, in particular to optoelectronic and visual information processing technology-based invasion detection. The technical problems needing to be solved are to detect and identify the invasion events of pedestrians or vehicles in an environment with poor light at night, and prevent false alarms. In the device, a modulated laser beam is used as a detection light source and is matched with a photoelectric signal collector, a photoelectric signal amplifier and a photoelectric signal processor to form a front-end unit; and a camera provided with active illumination and an embedded image processor are used as a rear-end response processing unit. When the invasion events occur in a detection area, a laser photoelectric signal is interrupted to trigger an active vision system so as to realize linkage and improve the reliability of the invasion detection. The laser-vision linked night invasion detection device can be used for anti-invasion security detection around residential areas, storerooms, airports and other important places.

The invention discloses a vehicle identification method based on fusion of radar and vision information. The vehicle identification method based on fusion of radar and vision information includes thefollowing steps: S100, entering all the information of a vehicle to be identified; S200, acquiring the current positional information and the motion state information of the vehicle; S300, acquiring the appearance information of the vehicle to be identified; S400, comparing the acquired appearance information of the vehicle to be identified with the entered vehicle information; and S500, recordingand uploading the information of the illegal vehicle and the vehicle driver. The vehicle identification method based on fusion of radar and vision information monitors the vehicles on the road in real time through the method of integrating the radar with vision information, determines the best position of shooting through the radar, can capture the clear image of the vehicle to be identified, canobtain the information of the vehicle to be identified through vision information processing, can determine whether the driver has illegal behavior through information comparison, and can greatly improve the accuracy and efficiency of vehicle identification through the method of combining the radar and the vision information.

The invention provides a visual information processing system and method. The system comprises a camera module, a control module and a calculation module connected with the control module through a high-speed data bus, wherein the camera module comprises one or more cameras used for collecting image information; the control module comprises an image receiving sub module used for receiving target image information collected from the one or more cameras, and an image preprocessing sub module used for performing image preprocessing on the target image information; and the calculation module comprises an image post processing sub module used for performing image post processing on the target image information subjected to the image preprocessing. Compared with the prior art, the visual information processing system and method have the advantages of realizing the integrated processing of multiple visual functions, ensuring more flexible visual information processing, having higher functionexpansibility and effectively reducing the equipment deployment cost on the whole.

The invention discloses a multichannel bionic vision method for recognizing a complex scene image, and the steps of the method mainly employ three transformation channels. For the first channel, the method comprises the steps: carrying out the forwarding channel calculation of an original image through simulating a multichannel vision processing mechanism of a human brain, obtaining a candidate target set, and generating the memory information at two stages at the same time; for a second channel, the method comprises the steps: sequentially carrying out the reverse channel calculation of all candidate targets through the memory information, and obtaining an impact image in the original image, so as to form a target edge image; for the third channel, finally carrying out the transformation of the target edge image for two times through employing a second forwarding channel algorithm, obtaining a feature transformation image, and carrying out the verification in the candidate set, so as to complete the image recognition. The method extracts the vision memory information through simulating the multichannel vision processing mechanism of the human brain, effectively simulates the multichannel vision processing mechanism of the human brain through the reverse channel calculation, and is high in accuracy of an obtained recognition image.

The invention discloses a spatio-temporal context target tracking method based on a human brain memory mechanism. The method introduces a visual information processing cognitive model of the human brain memory mechanism to the update process of a space-time relation model of the STC method, so that each template is allowed to be subjected to transmission and processing of transient memory, short-term memory and long-term memory spaces, and memory-based model update strategies are formed; by memorizing previous scenes, the method can still keep continuous robust tracking when problems of illumination variation, attitude abrupt change, shielding and reappearing after brief disappearance of a current target occur; besides, when a confidence map is calculated according to spatio-temporal context information, the method sets N target center location candidate points and selects a target center location, the similarity between which and a target template is maximum, as a final tracking result, thereby reducing errors caused by the confidence map, and improving tracking precision; and finally, a high-precision and strong-robustness moving target tracking method is formed.

The invention relates to a visual acuity automatically recognized unattended event related potential brain-computer interface method. The method comprises the following steps: manufacturing a vision-sighting target stimulus picture and storing the picture in an event related potential processing device; connecting the processing device with a multi-lead electrode cap for detecting an electroencephalogram signal; setting a visual stimulus parameter of the sighting target stimulus picture in the processing device to a user; exerting auditory stimulus on the user while performing vision stimulus on the user; acquiring an original electroencephalogram signal of the vision unattended event related potential of the user in the above mentioned process by use of the multi-lead electrode cap; processing and analyzing the original electroencephalogram signal by use of the processing device to obtain an event related oscillogram educed by different visual acuity sighting target stimulus users; establishing the relevance of the electroencephalogram and the visual acuity according to the oscillogram, and then establishing a visual acuity automatically recognized brain-computer interface. The invention provides a new method for the brain-computer interface for vision information processing and the automatic detection of the visual acuity, and the defect that the subjective coordination of the user is required is overcome.

The invention discloses a visual information fusion system and method for vehicle positioning. The system comprises a vehicle-mounted visual sensor module, an online visual information processing module and an offline training module. The output end of the vehicle-mounted visual sensor module is in signal connection with the input end of the online visual information processing module and the input end of the offline training module. The offline training module is used for training the proposed deep neural network and copying the network parameters obtained in the training stage to the onlinevisual information processing module. According to the invention, the image information acquired by the plurality of visual sensors in the vehicle-mounted visual sensor module can be effectively utilized, the problem that a positioning system cannot work normally under the condition that a single sensor fails is avoided through the complementation of the image information, and the vehicle positioning precision and the algorithm reliability are improved.

An illustrative mobile device includes a data storage configured to at least temporarily store visual information and at least one processor that is configured to determine whether to request visual information processing from a network with which the mobile device may communicate. The processor is configured to determine a mobile device condition and a network condition. The processor determines a type of feature from the visual information to use for classification based on the determined mobile device and network conditions. The processor is configured to classify the visual information based on the determined type of feature and determine a confidence indicator based on the classification. The processor determines whether to request visual information processing from the network based on the determined confidence indicator.

The invention relates to the field of structural design of visual information processing chips in integrated circuits, and discloses dual-bus visual processing chip architecture. The architecture comprises a first bus, a second bus, a visual computing and deciding module connected to the first bus, a first memory connected to the first bus, a characteristic combination and mode generating module connected to the second bus, an image characteristic extraction module connected to the second bus, a second memory connected to the second bus, and a bridge circuit for connecting the first bus and the second bus.

The present invention discloses a moving target tracking method based on the KCF and the human brain memory mechanism. A visual information processing cognitive model based on the human brain memory mechanism is introduced into the target template updating process of the KCF algorithm, and a method based on MTU (Memory-based Template Updating) forms a new model updating strategy, so that each template will be subject to transmission and processing of three spaces of instant memory, short-term memory and long-term memory. In the target tracking process, the target template is updated with different updating strategies according to the matching degree between the target template in the current frame and the target template in the memory space. By effectively remembering the previous scene, the method can keep the robust tracking even when the current target has the sudden change in attitude, reoccurs after temporary disappearance, is sheltered, or the like.

A biological information processing system includes processor including hardware. The processor performs: acquisition processing in which pulse wave information is acquired; processing in which analysis processing on the pulse wave information is performed to generate analysis result information; and output processing in which the analysis result information that is generated is outputted. The processor generates information for displaying distribution information of a pulse interval found on the basis of the pulse wave information and a reliability index of measurement of the pulse wave information, as the analysis result information.

The present invention discloses a real-time monitoring system for an elevator or an escalator, the elevator or escalator includes a control system arranged in a control cabinet, the control system includes a screen configured to display control information, the real-time monitoring system is arranged in the control cabinet, and includes: a camera configured to capture an image of the screen in real time; a visual information processing board configured to recognize the captured image as corresponding text data, and process the text data to form a message to be transmitted; and a data transfer unit configured to transmit the message from the visual information processing board to a cloud, wherein the cloud determines whether the elevator or the escalator needs maintenance according to the message.

The invention discloses an information interface layout beauty evaluation method based on cognitive characteristics, includes the following steps: Firstly, the mapping relationship between layout andcognition of information interface is constructed according to the visual information processing process and visual attention principle, Secondly, combined with the construction logic of quality evaluation model in software engineering field, the measurement index is deduced and the beauty evaluation model of information interface layout is established. Finally, the weight of each measurement index is calculated by AHP, and the comprehensive evaluation method of information interface beauty is constructed. The invention has high universality and provides the characteristics of a set of objective quantitative evaluation methods for the optimization of the layout of the information interface.

The invention discloses a networked multi-robot system. The system comprises: embedded visual nodes, a console and a multi-robot group. The embedded visual nodes are used to collect and process visual information and send processing results of the visual information to the console and the multi-robot group. The console, combining the visual information processing results of the embedded visual nodes and position and state information of the multi-robot group, is used to provide instruction information and environment information for the multi-robot group. The multi-robot group is used to send the position and state information of the multi-robot group itself to the console, receive the processing results of the embedded visual nodes to the visual information and the instruction information and the environment information of the console, combine perception of the multi-robot group itself and carry out a movement decision. In the invention, a plurality of the embedded visual nodes independently collect and process the visual information so that distributed parallel processing can be realized. The information can be timely provided for the multi-robot system. Decision quality and efficiency can be increased.

Embodiments of the present application provide a visual attribute recognition method and device and a medium. The method includes acquiring basic visual information of a target image by using a firstpart of a preset neural network model; using the second part of the preset neural network model to obtain the global visual information and the local visual information; according to the importance ofeach feature in the local vision information, processing the local vision information to obtain the first local adjustment information; according to the importance of each feature in the global vision information, processing the global vision information to obtain the first global adjustment information; adjusting the global vision information according to the first local adjustment information to obtain the adjusted global vision information; adjusting the local vision information according to the first global adjustment information to obtain the adjusted local vision information, and performing the visual attribute recognition on the target image according to the adjusted global vision information or/and the local vision information. The accuracy of visual attribute recognition can be improved.

The invention relates to an interlocking logic relationship visual information processing method based on a knowledge graph. According to the method, an interlocking logic expression relation stored in a text file is read through a python language, a triad module is extracted, triads are variables, relations among the variables and variables, the triads are stored in a graphic database, and an interlocking logic relation graph database is formed and used for inquiring the interlocking variables and logic relations. Compared with the prior art, the mapping knowledge domain technology is used inthe railway signal industry for the first time and is applied to visual expression of the interlocking logic relationship in the interlocking system for the first time, and reference is provided forapplication of the interlocking system and other aspects of the railway signal industry.

A monitoring device includes an imager, a processor and transmission media. The imager is used to capture visual information pertaining to a flame produced by an appliance. The processor processes the visual information to produce processed information. The transmission media is used by the processor to forward the processed information to a monitoring system.