The invention relates to a GPU image matting method. The method is characterized in that a CPU transmits foreground image data and background image data needed by image matting to a GPU global memory; the GPU carries out image matting processing and then passes back to the CPU; and through GPU multithread processing and polymerization, reading and optimization are performed. In the invention, parallel calculating of the GPU and a floating-point calculation capability advantage are used in image matting processing so that a requirement of the image matting to the CPU is reduced; performance is greatly increased and a requirement of a high definition signal can be satisfied.

GPU image matting method

The invention relates to a CUDA-based monitoring video foreground extraction acceleration method, which is applied to a graphic processing apparatus comprising a CPU and a GPU. The method comprises the steps that the GPU performs foreground processing on a video frame according to a background model to obtain foreground information; and the CPU obtains the foreground information of a video frame output in the GPU, performs correction on the background model according to the foreground information of the video frame, and applies the corrected background model to foreground processing of the GPU to the next video frame. According to the method, a high-performance computing capability of the GPU is combined with a branch operation capability of the CPU, hardware resources are fully utilized, the operation speed is greatly increased, and the efficient extraction processing demand of a video foreground is met.

CUDA-based monitoring video foreground extraction acceleration method

The invention discloses a background modeling method for long-time background collection, and the method comprises the following steps: 1, extracting a first frame of a video sequence as an initial image matrix initMat, the size of the initMat being width * height, and the unit being a pixel; step 2, initializing a background model according to initMat; step 3, sequentially reading frames in the video sequence, and performing background modeling; and 4, outputting a mask matrix and a bk matrix, the pixel point with the mask median value of 255 being the pixel point of the moving object, 0 being the background pixel point, and the bk matrix being the background model. The method has the advantages that foreground and background in a video can be separated from each other at the beginning ofa video sequence in a video sequence, and after background pixels are collected for a long time, the background modeling effect can be achieved.

Background modeling method for long-time background collection

The invention provides a face access recognition method. The method comprises the following steps: capturing a picture to obtain a to-be-verified picture; performing face recognition on the to-be-verified picture to obtain a face similarity score; calculating a safety score (Score) according to the face similarity score (Scorelike), popularity coefficient (Csafe) and crowd risk coefficient (Ccrowd), wherein the Score is Scorelike*Ccrowd*Csafe; and controlling the access according to the safety score Scorelike. The access method is dynamically adjusted according to a risk control strategy whena user uses the face access. The safety and convenience of the access are taken into consideration by combining the advantages of face access with fingerprint, iris and other assisted access methods.

Face access recognition method, system, computer storage medium and equipment

The invention discloses an emergency button compatible access control system and a positioning method. The system comprises an identification module, a key control module, a rotating module, an alarmmodule, a delayed closing module, a detection module, a card reader unit, an information reading unit, a data center checking unit, an information feedback unit, an information transmitting unit, an emergency button switch unit, an information converting and transmitting unit, an information receiving and processing unit, a variable speed motor unit, a rotating shaft, a door plate, an electric lock switch, a travel switch, an alarm, an infrared detection unit, an electromagnet, a control switch, a time control unit, an angle measurement unit, a controller and an information transmission unit.The rotating angle of the rotating shaft is detected through the angle measurement unit, whether the door plate rotates in place or not can be checked by checking the angle measurement unit, and whether the door plate is opened or closed can be known conveniently.

Emergency button compatible access control system and positioning method

The invention discloses a GPU adaptive foreground extracting method which is applied to a computer with a GPU image processor The method is characterized by comprising the steps of transmitting image data from an inner memory to a CPU; performing mixed gaussian background modeling for each pixel point corresponding one thread in the CPU, wherein the gaussian number in each pixel point can be dynamically adjusted; performing binaryzation for the maximum Gaussian weight in each pixel point; transmitting the treatment result as the final result to the inner memory through a global memory With the adoption of the method, the foreground extracting efficiency can be greatly increased; the movement target can be accurately extracted on real time in order to adapt to the actual environment well

GPU adaptive foreground extracting method

The utility model relates to a multistage intelligent entrance guard system that do not to select unblock more based on fingerprint identification, including computer, card reader, controller, the computer is connected or signal connection with the controller electricity, still includes fingerprint machine, electro -magnet, electric lock, alarm system, control button, opening of control button control gate, wherein the controller is connected or signal connection with fingerprint machine, electro -magnet, card reader, electric lock, control button, alarm system electricity respectively, and card reader, control button, the electricity connection of each other of electro -magnet difference or signal connection, the utility model provides a compatible access control system of the multiple mode of unblanking makes things convenient for the staff to pass in and out, has avoided the risk that quilts such as magnetic card, key usurped simultaneously.

Multistage intelligent entrance guard system that do not to select unblock more based on fingerprint identification

The utility model provides a become power station monitoring device based on panorama culmination intelligence video monitoring, includes that the video absorbs module, ARM embedded system module, teletransmission control module and man -machine interaction module, the video absorbs module, teletransmission control module all to be connected with the ARM embedded system module, and the ARM embedded system module is passed through the teletransmission module and is connected with man -machine interaction module, the video absorbs the module to comprise a plurality of gun type infrared camera machine, and a plurality of gun type infrared camera machines set up a plurality of eminences that occupy in the transformer substation respectively. The utility model discloses a become power station monitoring device based on panorama culmination intelligence video monitoring can realize the control and the high definition demonstration function in whole key place, and the installation is simple, only need realize macroscopic view control and object detection at selected rifle bolt of culmination installation, fixes a position and discerns the target, compares through the head portrait with staff's record, judges whether be the stranger, reports to the police, and the intellectuality of realization transformer substation is on duty.

Become power station monitoring device based on panorama culmination intelligence video monitoring

Popular Tasks used by Researchers

How can I help with your research?

Get more enhanced response by uploading the PDFs you want me to reference.

No relevant PDFs in your library

SciSpace is the AI research assistant for academics. Run systematic literature reviews on 280M+ papers, and write papers with cited sources. Trusted by 1M+ students, PhDs & researchers.

SciSpace | AI for Research

We encountered an error while loading the panel. Please try refreshing the page.

Something went wrong

Analyze PDFs

Code & Manuscripts

Funding & Grants

Literature & Patents

Medical & Clinical Data

Systematic Review

Visualize & Present

Web & Data

Build a Google Scholar-like website for your research.

Build a website

Create charts and images for your research

Create a Chart

Write a paper for submission to a journal

Draft a manuscript

Patent Search

Design eye-catching scientific posters in minutes.

Scientific Poster Generation

Systematic Literature Review

Anyone with the link can view this chat. You can switch it back to private anytime.

One task is running at the moment. Your messages will be shown right after.

Drag and drop or click here to browse

Loved by <highlight>1 million+</highlight> researchers

Extract a list of specific topics and their sources from unstructured text

Topics

Compare and analyze relevant papers that matches with your search

Papers

Get insights from PDFs and bookmarked papers from your library

My library

Recent searches

Try searching for:

Catch AI-generated content in scholarly and non-scholarly content

{ai} Detector

Ai Writer

Get PDF Summaries, highlighted text explanations 

Chat with PDF

Effortlessly create in-text citations and bibliographies in APA and 2,500 other formats

Citation generator

Get explanations, summaries, and answers on academic papers

Ease up your research workflow with {scispace}'s cohort of exciting AI tools

Elevate your academic writing skills and convey your ideas the way you want

Paraphraser

Explore our range of reading and writing tools

Your file is being prepared and should be ready in a few minutes. If it's a large file, it might take a bit longer. You can close this window, and we'll email you the file when it's done.

You have reached a maximum limit of <strong>{limit}</strong> columns in the table. Remove at least <strong>1</strong> column to add or create another one.

Dai Xianyu

Author Tools

Chat about Author

Papers

GPU adaptive foreground extracting method

Multistage intelligent entrance guard system that do not to select unblock more based on fingerprint identification

Become power station monitoring device based on panorama culmination intelligence video monitoring