21results about How to "Reduce computing power consumption" patented technology

The application discloses a same-row relationship detection method and device, electronic equipment and a storage medium. The method comprises the following steps: obtaining target video data in a target place; determining a personnel pixel area in each image frame in the target video data, and determining a target height and / or a target body shape ratio of the personnel corresponding to the personnel pixel area; determining an anonymous identity of the personnel in the image frame based on a dynamic anonymous personnel name list corresponding to the target place and the target height and / or the target body shape ratio; and judging whether a first search personnel and a second search personnel are in a same-row relationship according to the anonymous identity information of the personnel in each image frame in the target video data. The scheme can efficiently and with low algorithmic consumption detect the anonymous identity of the personnel in each image frame in the video data, so as to accurately and quickly find out whether the two search personnel are in a same-row relationship according to the anonymous identity of the personnel.

This invention belongs to the field of power monitoring technology and discloses a brightness balancing and regional smoothing fusion algorithm for power monitoring images. The method includes: image denoising, using mean filtering and median filtering to smooth the image; image registration, performing optimal matching on two or more images obtained from different sensors, at different times, or from different angles; and image region fusion, smoothly fusing images with different angles and brightness contrasts. This method is applicable to panoramic video monitoring systems. Designed specifically for the complex lighting and high electromagnetic interference characteristics of power operation sites, this invention effectively eliminates on-site noise through adaptive preprocessing, and achieves a balance between global brightness uniformity and local detail preservation through layered brightness balancing, solving the problems of incomplete brightness balancing and detail loss in power scenarios encountered by general algorithms.

The application relates to the technical field of intelligent diagnosis of power supply systems and optical communication cross technology, and discloses a power supply system health diagnosis method based on adaptive multi-dimensional ripple feature fusion, which comprises the following steps: short-time coarse sampling of a power supply ripple signal, dynamic adjustment of sampling parameters and window function types according to the coarse sampling result; extraction of frequency domain features, time domain features and time-frequency domain features to form a multi-dimensional feature vector; dynamic updating of a fault threshold value through Bayesian estimation based on historical normal feature samples in a fault feature sample library; adoption of weighted D-S evidence fusion of the fault confidence of the multi-dimensional feature vector based on the fault threshold value, and determination of a system state based on the fault confidence to output a diagnosis result; and hierarchical fault tracing when the system state is an abnormal state. The application realizes adaptive sampling and feature fusion, reduces the fault misjudgment rate, shortens the tracing time consumption, has low computing power consumption, and is suitable for real-time health diagnosis of a data center optical module power supply system.

This application discloses a method, system, device, and computer medium for processing vehicle lateral speed, relating to the field of vehicle control technology. The method involves obtaining a sampling time interval for the vehicle lateral speed, generated based on computational and noise limitations for lateral control; obtaining an update time interval for the vehicle lateral speed, where the update time interval is less than the sampling time interval; determining the update moment for the vehicle lateral speed based on the update time interval; and generating the latest vehicle lateral speed according to the sampling time interval when the current moment matches the update moment, so as to perform lateral control based on the latest vehicle lateral speed. The lateral speeds generated in two consecutive iterations are independent, resulting in smooth and continuous lateral speeds. This allows for higher-frequency updates of the vehicle lateral speed, reducing computational consumption and noise impact, ensuring accurate lateral control of the vehicle even under computationally limited conditions.

This disclosure provides a method, apparatus, device, medium, and vehicle for predicting traffic light switching points, relating to the field of data processing technology, specifically big data, intelligent transportation, and cloud computing technologies. It can be applied to navigation, vehicle-to-everything (V2X) technology, smart cockpit, and autonomous driving technology. The method includes: acquiring target intersection image data; performing image recognition on the target intersection image data to obtain a target traffic light identification result; and sending the target traffic light identification result to a server. The server calculates the switching point information of the target traffic light based on the target traffic light identification result and the target traffic light's switching pattern information. This disclosure reduces the time and cost of traffic light switching point prediction, as well as the computational power consumption of the server, thereby improving the accuracy and efficiency of traffic light switching point prediction.

The application discloses a kind of based on efficient reinforcement fine tuning and rule check multi-modal industrial intelligent detection method and system, method includes: S1.to industrial detection data is divided, and reference sample library and industrial acceptance rule base are constructed;S2.in frozen multi-modal base model, introduce interlayer collaborative adjustment module, and build feature space;S3.in feature space, extract locatable abnormal evidence, and output structured detection result that can be directly accessed in industrial acceptance rule base and check;S4.structurized detection result is input to industrial acceptance rule base and is checked, obtains industrial rule check result;S5.based on industrial rule check result, the related parameter of interlayer collaborative adjustment module is reinforced fine tuning, obtains multi-modal detection model;S6.based on multi-modal detection model, the abnormal detection and industrial acceptance determination of sample to be detected on production line are carried out, obtains final detection result.

This invention discloses a high-precision RTK positioning method based on an improved adaptive genetic algorithm-based satellite selection strategy, belonging to the field of satellite navigation and positioning technology. A subset of satellites with a specified number of satellites is selected to minimize the loss of the Position Accuracy Factor (PDOP) and satellite geometric volume. The satellite selection problem is transformed into a combinatorial optimization problem. A fitness function is designed with the reciprocal of PDOP and geometric volume as its core, and adaptive crossover and mutation probabilities are introduced. The genetic operation parameters are dynamically adjusted according to the population evolutionary state to balance global search and local convergence capabilities. Iterative evolution is performed under the condition of satisfying subset size constraints. When the convergence condition is met, the subset of satellites with locally optimal PDOP values ​​is output for subsequent RTK ambiguity resolution and positioning. This invention can quickly converge to a near-globally optimal satellite geometry configuration in polynomial time while also considering satellite quality, significantly reducing computational complexity and improving the real-time performance and reliability of RTK positioning.

This invention relates to the field of industrial manufacturing information management and data processing technology, specifically to a method and system for integrating and querying piecework data related to departmental relationships through tree-like node linkage. The method includes: constructing a hash-dependent directed acyclic graph (DAG) corresponding to the topological structure; obtaining the current composite fingerprint of the target root node; retrieving the current composite fingerprint from a cache library; if the retrieval is successful, directly returning the pre-stored aggregation result mapped in the cache library; otherwise, traversing downwards along the hash-dependent DAG, determining if the fingerprint has not changed and directly calling the pre-stored aggregation result; for sub-branches without a composite fingerprint mapping in the cache library, determining if the fingerprint has changed and performing real-time data aggregation calculation to generate a new aggregation result; establishing a mapping relationship and writing it into the cache library for subsequent query calls. This invention ensures millisecond-level real-time response of piecework cost data in scenarios with frequent engineering changes, eliminating data lag between design changes and financial accounting.

This invention relates to the field of IoT communication technology for lubrication equipment, and discloses a method and system for switching control of IoT edge communication modes for lubrication equipment. Edge nodes collect real-time communication environment data, equipment operation data, and business data requirements. Based on this multi-dimensional data, a multi-dimensional switching decision model is constructed. Physical dimension differences are eliminated through standardization processing, and a comprehensive score for each supported communication mode is dynamically calculated. Dynamic monitoring is performed based on the comprehensive score and trigger conditions including preset difference parameters. When the conditions are met, the target communication mode is selected and seamless switching is performed. When a device fault is detected, a fault adaptation evaluation value is calculated, and a forced switch to a low-latency communication link is initiated. This invention achieves deep coupling between the underlying communication state and mechanical operating conditions, thereby possessing edge emergency takeover and overflow prevention buffering capabilities under network-wide interruption, ensuring equipment operational safety.

This invention provides a method and system for road condition data acquisition and analysis based on a commercial vehicle connected driving recorder. The invention utilizes a connected driving recorder, a mandatory standard feature in commercial vehicles, to acquire road images in real time via an image acquisition module and perform preprocessing such as backlight compensation, distortion correction, noise reduction, and effective area cropping. An image recognition module employs a lightweight deep learning and rule-based algorithm to locally identify congestion and traffic accidents. A positioning module acquires latitude and longitude, vehicle speed, and timestamps via GPS / BeiDou dual-mode, achieving precise spatiotemporal binding through timestamp alignment and interpolation compensation. A 4G communication module uploads data using a strategy prioritizing event triggering and supplementing with periodic heartbeats. A data monitoring platform verifies the data, performs 3D deduplication, and then visualizes it on an electronic map. This invention requires no additional hardware, is low-cost, has wide coverage, and strong real-time performance, enabling automatic and precise monitoring of road conditions across the entire area, and is suitable for traffic management and logistics scheduling.

The invention discloses a metal plate strip foil surface defect online visual detection method. The method comprises the following steps: acquiring a metal surface defect image through multi-angle imaging and bright and dark field composite illumination; performing background adaptive preprocessing and feature enhancement on the defect image; a multi-scale convolution structure is adopted to extract global features, and spatial features are reconstructed through feature fusion and lightweight processing; triggering region-of-interest extraction based on defect saliency judgment, and performing loop feature enhancement on a local region; the defects are classified and graded by using a lightweight convolutional neural network, and the model is dynamically updated through online learning and a parameter adaptive optimization mechanism. The method is suitable for online detection of the surface defects of the metal plate strip foil in high-speed operation, the detection rate and classification accuracy of the surface defects of the tiny and low-contrast metal plate strip foil are remarkably improved while the real-time performance is guaranteed, and the detection precision of the surface defects of the metal plate strip foil is improved; and real-time feedback and quality tracing of the surface defect detection result of the metal plate strip foil are realized.

The application discloses a memory-compute FPGA architecture. Memory-compute cores, memory blocks and logic function blocks used for replacing part of the logic function blocks are arranged in a row-column array and arranged in an island type architecture. In the memory-compute cores, there are SRAMs arranged in a row-column array, the SRAMs store one-bit data, the output ends of an address decoding circuit are connected with each row of the SRAM array, the output ends of a write control circuit are connected with each column of the SRAM array, each eight columns of the SRAM array and an input register are connected with an 8-bit input port of an addition tree through a multiplexer and eight NOR gates, and the output port of the addition tree is connected with an output register through a shift and addition module and a quantization module in sequence. The application realizes in-memory computation on the FPGA, utilizes the parallelism and reconfigurability of the FPGA to support mapping of various networks, and utilizes in-memory computation technology to reduce data movement and further reduce computation power consumption.

The application discloses a van der waals correction calculation method based on a heterogeneous computing architecture and related equipment, and belongs to the technical field of data processing. The method comprises the following steps: in response to a van der waals correction calculation task instruction, a standard data packet transmitted by a master processor is received; based on the standard data packet, a van der waals correction calculation process is performed in combination with on-chip core storage of a slave processor, and a van der waals correction calculation result is obtained; and the van der waals correction calculation result is returned to the master processor. The application realizes heterogeneous computing by adopting a master processor and a slave processor in cooperation with a hardware architecture design, clearly defines the task boundary of different processors, avoids resource redundancy and waste of computing power, and realizes integrated storage and calculation design by performing van der waals correction calculation in combination with on-chip core storage of the slave processor, breaks through the data carrying bottleneck of the traditional architecture, greatly shortens the calculation period of a large-scale system, and reduces the consumption of computing power.

The invention relates to a coagulant accurate adding system based on multi-source perception and AI-digital twinborn fusion, and relates to the technical field of intelligent control, and the coagulant accurate adding system comprises a microfluidic water quality detection module used for carrying out multi-parameter synchronous detection on raw water and outputting multiple paths of water quality voltage signals; the pulse signal generation module is used for converting the multiple paths of water quality voltage signals into multiple paths of pulse width modulation pulse signals; the pulse neural network decision module comprises an input layer, a coagulation response mapping layer, a flocculation effect prediction layer and a dosing amount output layer; the piezoelectric micro-jet execution module comprises a piezoelectric micro-jet array and a piezoelectric driving circuit; the feedback regulation module comprises an on-line turbidity sensor arranged at the water outlet end of the coagulating basin and is used for realizing closed-loop dynamic optimization of the adding amount of the coagulant; according to the invention, the adaptability of the coagulation process to the water quality fluctuation of raw water is effectively improved, the medicament consumption and the sludge yield are reduced, and the stability of the effluent quality is guaranteed.

This application discloses an adaptive obstacle detection method and apparatus for adverse weather conditions, relating to the field of autonomous driving technology. The method acquires multi-source sensor data from an onboard vehicle, combines it with pre-calibrated sensor parameters to generate spatiotemporally aligned multi-source sensor data, extracts features to obtain feature maps for each modality, and projects all feature maps onto a bird's-eye view BEV space of the same resolution, using the LiDAR coordinate system as a reference, outputting a multi-modal BEV feature set. The feature set is concatenated along channels to form global state features, which are input into a pre-trained dual-branch adaptive decision network to obtain the output decision result, thereby generating the final image BEV features. Combining the multi-modal feature confidence weight matrix, the final image BEV features and the multi-modal BEV feature set are weighted to obtain globally fused BEV features, which are input into a 3D obstacle detection head for decoding and output detection results. This application can improve the robustness of multi-sensor fusion perception and obstacle detection accuracy under adverse weather conditions.

This invention relates to a low-power edge-side sleep staging method and system based on a liquid state machine in the field of sensing and computing technology. The method includes the following steps: inputting a single-modal target inference signal into a solidified liquid state machine training model; using the single-modal target inference signal to drive the LSM reservoir of the solidified liquid state machine training model to update the internal state of neurons and generate a membrane potential state vector; based on the membrane potential state vector, generating dynamic gating coefficients for adjusting the internal recurrent feedback loop of the LSM reservoir through the feedforward network calculation of the cross-modal dynamic gating module of the solidified liquid state machine training model; and performing weighted adjustment based on the dynamic gating coefficients and the recurrent feedback current of the LSM reservoir, outputting the current sleep staging result through the readout layer of the solidified liquid state machine training model. This solves the problem of high computational load and high power consumption when performing signal processing on resource-constrained edge devices in existing sleep monitoring.

This invention discloses an AI-powered holographic image adaptive enhancement system that adapts to network latency fluctuations, belonging to the field of communication and holographic imaging technology. By monitoring network latency, bandwidth, and packet loss rate in real time, this invention dynamically and adaptively adjusts the resolution, frame rate, data compression ratio, and transmission priority of the holographic image using AI algorithms. It is compatible with all network types, including 6G, 5G, WiFi 6 / 7, and Ethernet. When network conditions are good, it improves image quality; when network fluctuations occur, it prioritizes ensuring smooth interaction, achieving an optimal balance between image quality and latency. Simultaneously, it achieves multi-dimensional synchronous adaptation of holographic images, spatial audio, and interactive commands. This invention significantly avoids image stuttering, tearing, and blurring caused by network fluctuations, while reducing terminal computing power and power consumption, and can be widely adapted to various holographic interactive terminals.

The present application relates to a kind of power quality transient event capture method, the method comprises the following steps: step A, system initialization process;Step B, cycle sampling judgment process;Step C, state machine judgment process.Compared with prior art, the present application has reduced the calculation power consumption required for transient event capture and the corresponding hardware cost and other advantages.

The application relates to the field of image rendering, in particular to a 3D model cutting method, a rendering method, a cloud server, a client and a medium. The method comprises the following steps: obtaining model cutting data corresponding to a target model, wherein the model cutting data comprises a cutting range and vertex data, and the vertex data is vertex data corresponding to each vertex included in the target model; performing cutting calculation on the target model based on the model cutting data corresponding to the target model, to obtain a cutting calculation result, wherein the cutting calculation result is used for representing model information in the target model within the cutting range; and sending the cutting calculation result to a client, so that the client performs model rendering through a GPU. The application has the effect of reducing GPU computing power consumption.

The application discloses a video processing method and device, equipment and storage medium, and relates to the technical field of video, which comprises the following steps: acquiring a current bandwidth usage rate, and writing a to-be-analyzed video stream collected by a front-end camera into an edge storage device if the current bandwidth usage rate reaches a preset threshold; analyzing a time period of a current task according to a preset database to determine a current bandwidth idle time period, wherein the preset database comprises historical bandwidth usage data in a preset time period; determining a task execution time window according to the current bandwidth idle time period and a computing power resource utilization rate; and performing video processing on the corresponding to-be-analyzed video stream in the edge storage device based on the task execution time window. Compared with the existing real-time processing architecture video analysis method, the application can be optimized and configured through the bandwidth idle time period, the bandwidth occupation and computing power consumption in the peak period are reduced, and the resource utilization efficiency is improved.