Geologically constrained infrared imaging detection method and system for urban deeply-buried strip-like passage

    公开(公告)号:US12106024B2

    公开(公告)日:2024-10-01

    申请号:US18159922

    申请日:2023-01-26

    Abstract: Provided in the present invention are a geologically constrained infrared imaging detection method and system for an urban deeply-buried strip-like passage, pertaining to the crossing fields of geophysics and remote sensing technology. The method includes: establishing an urban hierarchical three-dimensional temperature field model according to urban street DEM data and geological data corresponding to urban streets; acquiring urban stratum geological background heat flux according to the urban hierarchical three-dimensional temperature field model; using a total solar radiation energy distribution model to calculate urban surface total solar radiation energy; sequentially filtering out the urban surface total solar radiation energy and the urban stratum geological background heat flux from an infrared remote sensing image of a region corresponding to a strip-like underground target, to acquire a perturbation signal image of an urban street deeply-buried strip-like passage; and using grayscale closed-operation plus an edge detection algorithm to perform detection and positioning after preprocessing the perturbation signal image of the urban street deeply-buried strip-like passage, to acquire location information of an urban strip-like underground passage. The present invention achieves inverse detection and positioning of an urban street deeply-buried strip-like passage.

    SAMPLE-DIFFERENCE-BASED METHOD AND SYSTEM FOR INTERPRETING DEEP-LEARNING MODEL FOR CODE CLASSIFICATION

    公开(公告)号:US20240192929A1

    公开(公告)日:2024-06-13

    申请号:US18475447

    申请日:2023-09-27

    CPC classification number: G06F8/35 G06F8/42

    Abstract: A sample-difference-based method and system for interpreting a deep-learning model for code classification is provided, wherein the method includes a step of off-line training an interpreter: constructing code transformation for every code sample in a training set to generate difference samples; generating difference samples respectively through feature deletion and code snippets extraction and then calculating feature importance scores accordingly; and inputting the original code samples, the difference samples and the feature importance scores into a neural network to get a trained interpreter; and a step of on-line interpreting the code samples: using the trained interpreter to extract important features from the snippets, then using an influence-function-based method to identify training samples that are most contributive to prediction, comparing the obtained important features and the most contributive training samples, and generating interpretation results for the object samples. The inventive system includes an off-line training module and an on-line interpretation module.

    PNEUMATIC SOFT DEXTEROUS HAND FOR PATIENT WITH MISSING FINGER FUNCTIONS AND SOFT ROBOT

    公开(公告)号:US20240189117A1

    公开(公告)日:2024-06-13

    申请号:US18324995

    申请日:2023-05-28

    CPC classification number: A61F2/586 A61F2/68 A61F2002/5038

    Abstract: A pneumatic soft dexterous hand for a patient with missing finger functions and a soft robot relate to the technical field of soft robots. The pneumatic soft dexterous hand includes a soft finger, the soft finger includes a rubber casing, a silicone tube, an endoskeleton, and a built-in air bag, and the silicone tube is sleeved in the rubber shell. The silicone tube is formed with a chamber, and the endoskeleton and the built-in airbag are arranged in the silicone tube. The endoskeleton includes a plurality of skeleton modules hinged together, the plurality of the skeleton modules are divided into two groups, and the skeleton modules of the two groups are hinged in an alternating manner in sequence from left to right. The built-in airbag is embedded on the endoskeleton to support the endoskeleton.

    Method and system for predicting junction temperature of power semiconductor module in full life cycle, and terminal

    公开(公告)号:US11976984B1

    公开(公告)日:2024-05-07

    申请号:US18182339

    申请日:2023-03-12

    CPC classification number: G01K7/22 G01K15/005

    Abstract: The present disclosure belongs to the technical field of power electronic converters, and discloses a method and a system for predicting a junction temperature of a power semiconductor module in the full life cycle and a terminal. The method includes the steps: arranging an NTC thermistor network to monitor the temperature of each area inside the power module when the power module works; obtaining data for training the neural network by utilizing finite element simulation or experiments, and building a neural network model among the temperature of the NTC resistor network, a water flow rate, an aging factor and the junction temperature of the chip under working conditions. The present disclosure improves the junction temperature prediction accuracy of areas with relatively large errors comprehensively and realizes the high-precision junction temperature prediction under all working conditions.

Patent Agency Ranking