14results about How to "Accurate interpretation" patented technology

The embodiment of the invention provides a video explanation generation method. The video explanation generation method comprises the following steps: acquiring audio and visual information of a target video; based on the audio and a pre-trained automatic speech recognition model, obtaining a recognition text; inputting the recognition text into a pre-trained language model to obtain a commentary text of the target video through the language model; and inputting the visual information and the explanation text into a pre-trained multi-modal model to obtain an explanation audio of the target video through the multi-modal model. According to the technical scheme, the automatic speech recognition model, the language model and the multi-modal model can be deeply integrated, the full-automatic process from the video to the high-quality explanation audio is achieved, the method is suitable for various complex explanation scenes, the features of different modals such as vision and texts can be effectively combined, and the intellectual property of the system is improved. The commentary audio which is more natural and smoother and highly matched with the video content is synthesized, and the audience experience is improved.

The invention relates to the technical field of intelligent adjustment, and discloses an environment adjustment system with emotional intelligence, which comprises a data acquisition module used for acquiring language signal data of a user, including language data and voice signal data; the language processing module is used for performing text preprocessing on the language data to obtain language keywords; the emotion score determining module is also used for determining a language emotion score of the user according to the emotion knowledge base and the language keywords based on the optimized language model; the voice processing module is used for performing feature extraction on the voice signal data, determining feature data of the voice signal data, and determining a voice emotion score of the user according to the feature data based on a voice model; and the environment adjusting module is used for determining a comprehensive emotion score of the user according to the language emotion score and the voice emotion score, and adjusting environment light configuration of the user according to the comprehensive emotion score. According to the invention, the emotion of the user can be sensed in real time and indoor light can be automatically adjusted according to the emotion so as to better meet personalized requirements of the user.

The invention provides a primer group and a kit for detecting NF1 gene mutation by using a long-read-long sequencing platform and application of the primer group and the kit. The primer group provided by the invention comprises primers for amplifying all exons of the NF1 gene and intron regions of flanks at two ends of the exons, and primers for amplifying the NF1 gene and deletion mutation of 50kb regions of the flanks at two ends of the NF1 gene. In combination with the characteristics of a long-read-long sequencing platform, various mutations of the NF1 gene can be comprehensively, accurately, quickly and simultaneously detected at high throughput.

The invention relates to the field of brain image processing, in particular to an image analysis processing system and method based on thought control, and the system comprises an electroencephalogram signal collection system which is used for collecting and preprocessing electroencephalogram signals of a user; the thought analysis system is used for extracting features from the preprocessed electroencephalogram signals and then carrying out thought classification and instruction mapping; the image acquisition and processing system is used for acquiring image information according to the mind control instruction of the user and performing image processing; the VR display system is used for converting the processed image into a format suitable for VR equipment display and providing interactive feedback; and the control and communication system is used for converting the mind state into a control instruction and realizing data transmission and communication among the systems. The EEG signals and the fNIRS signals are acquired through the multi-mode signals, the defect of a single signal in the aspect of reflecting the brain activity is overcome, and therefore brain activity information corresponding to the thought is more accurately interpreted.

This invention provides an intelligent inversion method for ground mechanical parameters driven by impact penetration test data. It constructs a machine learning inversion model by learning and training the relationship between the time history curves of the penetrometer's dynamic response and ground mechanical property parameters in the interpretation model database. After acquiring measured cone resistance and acceleration time history data, these are input into the trained machine learning inversion model to directly obtain the ground mechanical property parameters. Furthermore, requiring only one input-output reverse transmission, the inversion speed can be reduced to the second level. The intelligent algorithm enables accurate interpretation of ground mechanical property parameters such as soil density, elastic modulus, Poisson's ratio, cohesion, and internal friction angle. This invention is rationally conceived and can efficiently, accurately, and intelligently interpret the ground mechanical property parameters of impact-penetrated soil.

The application discloses a wireless photoelectric hybrid sensing-integrated control method and system based on a multi-modal large model, and belongs to the technical field of wireless communication and artificial intelligence. A distributed network composed of sensing-integrated nodes and decision coordination units is constructed, the sensing-integrated nodes are used to synchronously realize wireless sensing and communication functions, and the transmission capacity and reliability are improved in combination with a photoelectric hybrid link (radio frequency and wireless light). A lightweight multi-modal large model integrated with the nodes is used for fusion coding and deep understanding of multi-source data such as sensing signals, vision and point clouds, decision instructions are generated in real time at the edge side, the instructions are quickly issued and executed through the distributed network, and a "sensing-communication-decision-execution" closed loop is formed. Through a local model fine-tuning mechanism, model parameters and resource strategies are cooperatively optimized, and the method has the advantages of millisecond-level delay, high decision accuracy, high energy efficiency and strong robustness, and is suitable for scenes such as automatic driving and industrial Internet of Things that have strict requirements on real-time performance and reliability.

The invention relates to the technical field of data processing, and discloses a static application security detection false alarm judgment system based on LLM. Comprising the following steps: extracting security detection data, performing de-duplication processing on the replaced security detection data, calculating the similarity of the security detection data, constructing a vector library, calculating the real-time similarity of the security detection data to be discriminated, and screening matched data from the vector library. Interpreting annotation data corresponding to the to-be-discriminated security detection data according to the big language interpretation model; compared with the prior art, each replaced security detection data can be ensured to keep independence and uniqueness, the similarity calculation burden is reduced, the similarity calculation time is saved, and the similarity calculation efficiency is improved. And simple and understandable natural languages corresponding to the safety detection data can be accurately and quickly interpreted from the complicated and lengthy safety detection data, so that non-professionals can conveniently carry out false alarm discriminant analysis operation on the safety detection data of the application program.

The application relates to the technical field of image data processing, in particular to a geological remote sensing interpretation method and system for mineral resource exploration, which comprises the following steps: acquiring a geological remote sensing image to be analyzed in a mineral resource area; analyzing the uniformity of the distribution of the pixel gray value of all local areas in the image to be analyzed to obtain the image smoothness; acquiring the LBP value of all pixel points in the image to be analyzed and a historical standard sample image and performing clustering; using a seed growing algorithm to obtain each growth area of each image; determining an image detail evaluation index of the image to be analyzed through the overall distribution of the edge complexity of the growth area of each cluster to which the image to be analyzed and the historical standard sample image belong, obtaining an image enhancement effect quality evaluation, adjusting the wavelet threshold in combination with the image detail evaluation index, and interpreting the enhanced image. The application aims to improve the accuracy of geological remote sensing image interpretation in mineral resource exploration.

The invention provides a lithology remote sensing interpretation method and device, electronic equipment and a storage medium, and relates to the technical field of remote sensing interpretation, and the lithology remote sensing interpretation method comprises the steps: obtaining lithology geological data of a target region; the lithologic geological data are input into a trained lithologic interpretation model, lithologic label data are generated, and the lithologic interpretation model comprises an image text extraction module used for extracting image features and text knowledge features of the lithologic geological data; the graph network module is used for constructing a co-occurrence matrix according to the spatial co-occurrence relation of the target area, constructing a text knowledge matrix according to the text knowledge features, adding and fusing the co-occurrence matrix and the text knowledge matrix, generating an adjacent matrix, and generating a guide visual feature graph according to the adjacent matrix and the image features; and the interpretation module is used for decoding the guide visual feature map and generating lithologic label data. The lithology remote sensing interpretation accuracy and reliability can be remarkably improved, and therefore complex lithology scenes can be accurately interpreted.

This invention relates to the field of animal model preparation technology, and discloses a method for preparing a mouse model of brain hemorrhage without thermal effects. The method includes preparing a microbubble contrast agent and healthy adult mice, pre-treating the mouse brain and marking target sites, anesthetizing the mice and fixing them on a stereotaxic apparatus, aligning a hydrogel cone mounted on an ultrasound transducer with the target sites in the mouse brain, intravenously injecting the microbubble contrast agent, and 20-40 minutes after the injection, activating the ultrasound transducer to perform ultrasound intervention on the target sites in the mouse brain. The cavitation effect of ultrasound induces focal brain hemorrhage, establishing a mouse model of brain hemorrhage. This invention, by applying the hydrogel cone to the target sites in the mouse, allows sufficient time for heat dissipation from the brain tissue during the ultrasound pulse intervals, making the temperature rise of the target tissue negligible throughout the irradiation process. This effectively reduces thermal damage, leading to more accurate interpretation of experimental data and more reliable research conclusions.

The application discloses a precipitation phase state vertical detection and inversion method based on radio wave reflection and relates to the technical field of meteorological remote sensing detection. The application comprises S1, multi-frequency band radio wave synchronous transmission and echo signal collection: through a multi-frequency band radio transceiving system, multiple vertical detection radio waves of different frequency bands are synchronously transmitted, echo signals generated by precipitation particles in different height layers of the atmosphere are collected, and continuous collection of echo signals of full height layers is completed; through synchronous transmission of multi-frequency band radio waves and collection of echo signals of full height layers, the application obtains multi-dimensional scattering characteristic vertical profiles of different phase state particles, realizes decoupling division of phase state intervals of different height layers by constructing a multi-frequency band exclusive feature matrix and combining melting layer vertical gradient characteristics, corrects particle proportion weights in mixed phase state intervals by introducing adjacent height layer phase state continuity constraint conditions, and realizes accurate interpretation of the mixed phase state of the zero-degree layer melting zone.

The invention provides a model training method, a sound recognition method, equipment and a storage medium, and relates to the technical field of emotion recognition. The method comprises the following steps: firstly, acquiring sample animal information of a plurality of sample animals and label information of each piece of sample animal information; secondly, performing model training according to the sample animal information of the plurality of sample animals and the labeling information of each sample animal, and generating a cry recognition model; when the target animal makes a sound, a sound audio and an animal image of the target animal are obtained, feature extraction is carried out on the sound audio, and audio parameters of the sound audio are obtained; and finally, inputting the audio parameters and the animal image into a cry recognition model, and obtaining the emotion and cry meaning of the target animal output by the cry recognition model. According to the method provided by the invention, the emotion recognition accuracy of the human to the animal and the interactivity between the human and the animal are improved.

The invention relates to the technical field of government affair data, in particular to a policy document analysis system and method based on the big data technology, and the system comprises a data collection module, a data preprocessing unit, a natural language processing NLP engine, a machine learning analyzer, and a data storage module. The policy document analysis system and method based on the big data technology have the beneficial effects that the policy document analysis system and method based on the big data technology are realized by adopting advanced natural language processing (NLP) and machine learning technologies. The system can accurately identify and interpret key information, including technical terms and complex policy contexts, in the policy document, so that a more accurate analysis result is provided.

This invention relates to the field of bamboo fiber preparation technology and discloses a bamboo fiber debonding device, including a debonding machine body. A feeding platform is fixedly connected to the outer wall of the debonding machine body, and a nozzle is fixedly connected to the top of the feeding platform. A discharge platform is fixedly connected to the outer wall of the fiber debonding machine body on the side away from the feeding platform. A collection frame is slidably connected to the bottom of the inner cavity of the debonding machine body. When it is necessary to shorten the spacing of the three protruding teeth, the operator holds the handle and rotates the rotating rod to make the three protruding teeth gradually converge. When it is necessary to increase the spacing of the three protruding teeth, the rotating rod is rotated in the opposite direction to drive the first protruding tooth to slide in the opposite direction to make the three protruding teeth gradually disperse. Thus, the fiber debonding density can be adjusted by adjusting the spacing of the three protruding teeth. At the same time, the spacing of the three protruding teeth can also be adjusted according to the width of the bamboo strips being conveyed, so that the device can still perform precise fiber debonding when conveying multiple sets of bamboo strips at the same time, avoiding insufficient fiber debonding that cannot meet the needs of subsequent processing.