Patents
Literature
Patsnap Copilot is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Patsnap Copilot

35results about How to "Reduce memory space requirements" patented technology

Method and device for reducing the processing time of data in communication networks

The invention relates to a method and a device for improving the processing time of received data in packet oriented applications during a transmission via communication networks, in particular via an IP network and a mobile communication network such as the Global System for Mobile Communication (GSM), Universal Mobile Telecommunication System (UMTS) or General Packet Radio Service (GPRS). The received data packets of the second protocol layer, for instance, the RLP frames, are combined to form data packets of the first protocol layer, for instance to PPP frames, directly by the receiver in the second protocol layer. The completely and correctly combined data packets are then released to the first protocol layer. During the release of the data packets to the first protocol layer, a difference is made between two modes. The intra-flow mode is a mode, wherein all completely generated data packets are released, i.e. independent from whether the data packets belong to the same or to two different data flows. In contrast thereto, a difference is made in view of the inter-flow mode. In said mode only the completely generated data packets belonging to different data flows are released to the first protocol layer.
Owner:TELEFON AB LM ERICSSON (PUBL)

Clustering integration method for image data of X-ray films

The invention discloses a clustering integration method for the image data of X-ray films. The method comprises the steps of S01, pretreating the image of an X-ray film to obtain the data of the image; S02, obtaining a gray value (Gi, j) at each point of the image and storing obtained gray values in a gray value matrix G, wherein the gray value (Gi, j) represents the gray value of a point at the i row and the j column of the matrix; S03, conducting the clustering analysis and treatment on the gray value matrix G according to the k-means improved algorithm-based clustering integration algorithm or the hierarchical clustering-based improved algorithm; S04, conducting the integration operation according to the HGPA algorithm. According to the k-means improved algorithm-based clustering integration algorithm, the selection of k initial cluster centers is optimized. According to the hierarchical clustering-based improved algorithm, data are simplified during the data pretreating process. In this way, points of the same gray value in the matrix are divided in the same cluster, and the number of initial clusters is 256 at most. Therefore, the observation difficulty of X-ray films is lowered, and even exogenous foreign matters can be found out. As a result, the method facilitates the doctor diagnosing process.
Owner:广州越神医疗设备有限公司

Parallel compressed sensing GPU accelerated real-time imaging system and method

The invention discloses a parallel compressed sensing GPU accelerated real-time imaging system and method. The system comprises an optical unit (I), an electrical unit (II) and a rear-end data pipeline processing unit (III) which are subjected to parallel pipeline processing, wherein the optical unit (I) is used for collecting an optical signal to obtain a target image signal, and segmenting and modulating the target image signal and then sending the segmented and modulated target image signal to the electrical unit (II); the electrical unit (II) is used for performing parallel complementary compressed sensing imaging, completing parallel complementary measurement and sending low-resolution image data to the rear-end data pipeline processing unit (III); and the rear-end data pipeline processing unit (III) is used for realizing high-resolution image reconstruction and real-time display by using a GPU accelerated compressed sensing high-speed reconstruction algorithm. According to the invention, a parallel pipeline scheme of the optical unit, the electrical unit and the GPU accelerated reconstruction part is used, so that compressed sensing real-time acquisition of low-resolution images and real-time reconstruction and display of high-resolution images are realized.
Owner:NAT SPACE SCI CENT CAS

Breathing courtesy lamp control method and system, vehicle and storage medium

The invention discloses a breathing courtesy lamp control method and system, a vehicle and a storage medium. The method comprises the steps that when the vehicle receives an unlocking signal, a vehicle body controller outputs a breathing greeting signal to a front combined lamp and a rear combined lamp, and the vehicle body controller supplies power to a front steering lamp and a rear steering lamp at the same time within the time period of 0-T1; supplying power to the front steering lamp and the rear steering lamp is stopped within the time period of T1-T2; power is supplied to the front position lamp, the rear position lamp and a daytime running lamp at the same time within the time period of T2-T3; power is continuously supplied to the daytime running lamp within the time period of T3-T4, and a first PWM signal with a specific frequency and a duty ratio of 0-100% is output to the front position lamp and the rear position lamp; and a first PWM wave output by an automobile body controller is recognized through the driving module of the front combination lamp, and a second PWM signal with the same duty ratio is output to the LED light source module, so the filtering effect of a circuit on the PWM wave can be avoided, it is guaranteed that lamplight response is consistent with the PWM signals output by the automobile body controller, and a good breathing greeting effect is formed.
Owner:CHONGQING CHANGAN AUTOMOBILE CO LTD

Multi-core processor-based multivariate deep network model reconstruction method and device

The invention discloses a multi-core processor-based multivariate deep network model reconstruction method and device. The method includes: acquiring a video stream image collected by a camera; selecting a logical combination relationship, and determining the cascading relationship between each deep network model in the deep network module and the corresponding output action according to the logical combination relationship; loading the corresponding deep network model; Call the multi-core dynamic allocation management command to calculate the complexity of the loaded deep network model, allocate corresponding memory and predetermined number of core processors for each deep network model according to the complexity; input the collected video stream image into the corresponding deep network model ; Analyze the scene information in the video stream image according to the specified output information obtained after being processed by each cascaded deep network model, and execute the corresponding output action. The invention requires low memory space, flexibly combines various deep network models to form a building block development mode, and improves user development efficiency and interest.
Owner:武汉星巡智能科技有限公司

A real-time preprocessing method and storage medium for video surveillance data

A real-time preprocessing method and storage medium for video surveillance data, the method comprising: obtaining the minimum channel map, and calculating the downsampled dark primary color D down , after converting the video image into a grayscale image, sort in descending order according to the grayscale value of the pixels, select a certain proportion of high-brightness pixels, and use the average value as the global atmospheric light estimation A; for the downsampled dark primary color D down Perform filtering processing to enhance the local small value to obtain the optimized down-sampled dark primary color D filter ; to D filter Up-sampling to get the scattered light estimation of each position of the video, according to the preset fog imaging model, using the scattered light estimation S and the global atmospheric light estimation A to inversely solve the fog-free picture of the t-th frame. The present invention obtains the dark primary color of down-sampling, reduces the amount of calculation, improves the running time, and is filter Scattered light can be estimated by up-sampling in blocks, and fog-free images can be reversed by blocks, which reduces the demand for memory space and is more suitable for the situation of uneven equipment in the Internet of Things environment.
Owner:INFORMATION SCI RES INST OF CETC

Early Alzheimer's disease diagnosis system based on artificial intelligence

The invention discloses a method for diagnosing early Alzheimer's disease based on artificial intelligence. After acquiring text reading aloud recordings, it is automatically translated into text through an ASR program; and then the text is processed according to the following new algorithm of NLP to determine whether the user has AD. Compared with the traditional NLP algorithm, the new algorithm obtains the feature sets of the AD group and the healthy person group respectively, and performs set operations on the feature sets to achieve effective dimensionality reduction of the features, which not only improves the recognition accuracy of the model, but also reduces the accuracy of the algorithm. The demand for memory space speeds up the calculation speed. This method can be applied to computer terminals as well as mobile terminals. The user only needs to speak for a few minutes according to the prompts, without the intervention and analysis of professional doctors. The machine can give the judgment of whether the user suffers from AD through artificial intelligence. The method does not cause trauma to the human body, has low cost, fast diagnosis speed, and is easy to use. It is an effective technical means for large-scale screening of early AD.
Owner:CHANGSHU INSTITUTE OF TECHNOLOGY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products