Abstract: A charge-discharge method applied to an electronic device for controlling a charging and a discharging of a vehicle, the electronic device communicates with a charging pile. The charge-discharge method comprises collecting operation behavior information of a user with respect to household appliances, inputting the operation behavior information into a preset travel time prediction model to obtain a first driving travel time of the user, determining a first charge-discharge strategy of the vehicle of the user based on the first driving travel time, and transmitting the first charge-discharge strategy to the charging pile. The charging pile charges the vehicle or controls the vehicle to discharge based on the first charge-discharge strategy. An electronic device and a non-transitory storage are also disclosed.

Abstract: A multi-layer federated learning method based on distributed clustering is provided, which comprises the following steps. Computing a feature distribution similarity for each of participating nodes with non-(non-independent and identically distributed) data sets, and grouping these nodes into plural clusters by the feature distribution similarity. Updating local model of nodes of each cluster by a federated learning algorithm, and inputting these nodes into a multi-layer aggregation mechanism. Terminating the operation of the multi-layer aggregation mechanism until the clustering result meets a required demand. Furthermore, we implement a blockchain-based multi-layer federated learning system, including model aggregation module, API module, time-series synchronization module, and IPFS, based on distributed clustering architecture. The learning performance is proven to effectively improved.

Abstract: A medical data ownership management method comprises: Granting a first authority to a medical unit where the first authority allows the medical unit to generate a first medical object with its corresponding first non-fungible token (NFT) storing on a blockchain. Granting a second authority to a research unit, where the second authority allows the research unit to access the first medical object, and the research unit processes the first medical object to generate the second medical object with its corresponding second NFT storing on the blockchain. Finally, granting a third authority to a user unit, where the third authority allows the user unit to access the second medical object. The access control and ownership transfer of each medial object in the blockchain are achieved by trading the corresponding NFT and modifying its subordinate list. With the management method, the security and the privacy of the medical data will be enhanced.

Abstract: A method to obtain a more accurate and more immediate augmented reality (AR) user experience in relation to identifying feature-marked or sound-marked objects for sale for example obtains any initial data via a device and determines whether identifiable data is included therein using artificial intelligence (AI) techniques. If so determined, the initial data is stored. The method determining whether such identifiable data is comprised in a predetermined database. If yes, the identifiable data is identified by matching to obtain information. The information in the identifiable data can be stored to the predetermined database to retrain an AI identification model. After AI training of an identification model, data relevant to the commodity taken in written or audio can be outputted into the AR scenario. An electronic device and a non-transitory storage medium are also disclosed.

Abstract: An object recognition system is disclosed. The object recognition system includes an input device configured to obtain image data and a Radio Frequency Identification (RFID) data of an object, a processing device connected to the input device, and configured to perform a model training procedure, wherein the model training procedure includes capturing an object feature according to the image data, generating a classification data corresponding to the object according to the object feature, verifying a correctness of the classification data according to the RFID data, and generating a deep learning model according to the verified classification data, to regenerate the classification data, and an output device connected to the processing device, and configured to generate an object recognition data corresponding to the object according to the classification data generated by the deep learning model and the RFID data.

Abstract: A conference recording method is provided. The method includes obtaining a multimedia file corresponding to a conference, the multimedia file includes video data and audio data. Posture language of each person is recognized from the video data. Facial features and voice features of each person are extracted from the multimedia file. Personal identity information of each person is identified according to the facial features and the voice features of each person. Once the audio data corresponding to each person is converted into text information, the posture language, the personal identity information, and the text information corresponding to each person are output.

Abstract: An object recognition system is disclosed. The object recognition system includes an input device configured to obtain image data and a Radio Frequency Identification (RFID) data of an object, a processing device connected to the input device, and configured to perform a model training procedure, wherein the model training procedure includes capturing an object feature according to the image data, generating a classification data corresponding to the object according to the object feature, verifying a correctness of the classification data according to the RFID data, and generating a deep learning model according to the verified classification data, to regenerate the classification data, and an output device connected to the processing device, and configured to generate an object recognition data corresponding to the object according to the classification data generated by the deep learning model and the RFID data.

Abstract: A conference recording method is provided. The method includes obtaining a multimedia file corresponding to a conference, the multimedia file includes video data and audio data. Posture language of each person is recognized from the video data. Facial features and voice features of each person are extracted from the multimedia file. Personal identity information of each person is identified according to the facial features and the voice features of each person. Once the audio data corresponding to each person is converted into text information, the posture language, the personal identity information, and the text information corresponding to each person are output.

Abstract: An information transmitting method to identify information being transmitted while verifying the person giving such information comprises pre-storing feature image of at least one person and expression information corresponding to such feature image. An image of a scene is received and the image is analyzed for containing a preset body part. The preset body part is identified as being one of the feature images when the scene image contains a preset body part and expression information such as interest or non-interest and agreement or non-agreement indicated by the body part or face of the identified person is determined. An information transmitting system is also presented.