Abstract: The disclosure relates to an artificial intelligence (AI) system for mimicking functions, such as cognition and determination as of the human brain, by utilizing a machine learning algorithm such as deep learning, and an application thereof. Provided are a neural network learning method according to an AI system and applications thereof, the method including extracting, by using a masking filter having an effective value in a specific portion of the masking filter including weight information of at least one hidden layer included in a learning network model, characteristics of input data according to weight information of a filter corresponding to the specific portion, comparing output data with target data, the output data being obtained from the learning network model based on extracted characteristics of the input data, and updating a size of the specific portion having the effective value in the masking filter, based on a result of the comparing.

Abstract: The present disclosure provides a neural network training device for recognizing a class of an object included in an image based on an artificial intelligence (AI) system and an application thereof, the neural network training method including: acquiring, by using a first learning network model trained based on source training images respectively included in at least one class, feature information of a query image included in a class different from the at least one class; obtaining a generated image from the feature information of the query image by using a second learning network model acquiring feature information of the obtained generated image by using the first learning network model; and updating weights of layers respectively included in the first and second learning network models, based on a difference between the feature information of the query image and the feature information of the generated image and on a difference between the query image and the generated image.

Abstract: The disclosure relates to an artificial intelligence (AI) system for mimicking functions, such as cognition and determination as of the human brain, by utilizing a machine learning algorithm such as deep learning, and an application thereof. Provided are a neural network learning method according to an AI system and applications thereof, the method including extracting, by using a masking filter having an effective value in a specific portion of the masking filter including weight information of at least one hidden layer included in a learning network model, characteristics of input data according to weight information of a filter corresponding to the specific portion, comparing output data with target data, the output data being obtained from the learning network model based on extracted characteristics of the input data, and updating a size of the specific portion having the effective value in the masking filter, based on a result of the comparing.

Abstract: The present disclosure provides a neural network training device for recognizing a class of an object included in an image based on an artificial intelligence (AI) system and an application thereof, the neural network training method including: acquiring, by using a first learning network model trained based on source training images respectively included in at least one class, feature information of a query image included in a class different from the at least one class; obtaining a generated image from the feature information of the query image by using a second learning network model acquiring feature information of the obtained generated image by using the first learning network model; and updating weights of layers respectively included in the first and second learning network models, based on a difference between the feature information of the query image and the feature information of the generated image and on a difference between the query image and the generated image.

Abstract: The present invention relates to a side mirror for a vehicle, the side mirror being provided on a vehicle for facilitating the checking of road situations on the left, right, and rear sides. The side mirror for a vehicle comprises a mirror housing which is mounted on the exterior of the vehicle, and a mirror part which is supported by the mirror housing, and is divided into multiple mirror areas which are in aspherical form respectively across the horizontal direction from the internal side, which is close to the vehicle, to the external side, which is far from the vehicle. In the mirror part, the respective optical powers of the mirror areas gradually increase from the internal side to the external side so as to have the same optical power at the borders of the mirror areas, and the optical power rate for at least one of the mirror areas is constant.

Abstract: The present disclosure relates to a hardware protocol stack, and more particularly, to a hardware protocol stack to which a user-defined protocol is applicable. The present disclosure compares header information stored in a register unit with header information of a received frame, and determines a processing of the received frame on the basis of a comparison result. Also, according to a determined frame process method, data of the received frame is processed based on a logic according to the header information, it is stored at a designated position, or a response frame is transmitted.

Abstract: Disclosed is a PLC system having a plurality of CPU modules and a control method thereof, wherein the method includes ascertaining a clock signal when a count is a count corresponding to a time slot allocated by a master CPU module, generating a clock signal by accessing to a backplane, and ending generation of clock signal at a time when the access to the backplane ends.

Abstract: The present disclosure relates to a hardware protocol stack, and more particularly, to a hardware protocol stack to which a user-defined protocol is applicable. The present disclosure compares header information stored in a register unit with header information of a received frame, and determines a processing of the received frame on the basis of a comparison result. Also, according to a determined frame process method, data of the received frame is processed based on a logic according to the header information, it is stored at a designated position, or a response frame is transmitted.

Abstract: An access controlling method of a dual port memory system is provided. The method includes: requesting, by a first processor, an access from a dual port memory; and transmitting, by the dual port memory, a result signal according to the access request to the first processor, wherein the result signal includes a first result signal for notifying access success, a second result signal for notifying access failure, and a third result signal for notifying access hold.

Abstract: The present invention relates to a side mirror for a vehicle, the side mirror being provided on a vehicle for facilitating the checking of road situations on the left, right, and rear sides. The side mirror for a vehicle comprises a mirror housing which is mounted on the exterior of the vehicle, and a mirror part which is supported by the mirror housing, and is divided into multiple mirror areas which are in aspherical form respectively across the horizontal direction from the internal side, which is close to the vehicle, to the external side, which is far from the vehicle. In the mirror part, the respective optical powers of the mirror areas gradually increase from the internal side to the external side so as to have the same optical power at the borders of the mirror areas, and the optical power rate for at least one of the mirror areas is constant.

Abstract: An access controlling method of a dual port memory system is provided. The method includes: requesting, by a first processor, an access from a dual port memory; and transmitting, by the dual port memory, a result signal according to the access request to the first processor, wherein the result signal includes a first result signal for notifying access success, a second result signal for notifying access failure, and a third result signal for notifying access hold.

Abstract: Disclosed is a PLC system having a plurality of CPU modules and a control method thereof, wherein the method includes ascertaining a clock signal when a count is a count corresponding to a time slot allocated by a master CPU module, generating a clock signal by accessing to a backplane, and ending generation of clock signal at a time when the access to the backplane ends.

Abstract: The present invention relates to an apparatus and method for monitoring abnormal state of a vehicle. In the method, CAN data collected from an ECU mounted on the vehicle is transformed into coordinates. The coordinates are applied to a distribution map in a specific space, a number of clusters is calculated based on results of the application, and an initial center point corresponding to the number of clusters is selected. Clustering is performed based on the initial center point, and then clusters are generated. At least one piece of data is extracted from each of the clusters, and a state feature of a corresponding cluster is decided on using a difference between maximum and minimum values of attributes constituting the at least one piece of data. A current state of the vehicle is monitored based on current CAN data of the vehicle and state features of the clusters.

Abstract: Disclosed is an adaptive multi-redundant ring network system using a 2 port Ethernet communication module capable of selecting a path, the network system including a main network system including a first main control unit and a plurality of first slaves; and a sub-network system including a second main control unit and a plurality of second slaves, wherein each of the plurality of first slaves and each of the plurality of second slaves include 2-port Ethernet communication modules, and each of the plurality of first slaves, each of the plurality of second slaves, the first slave and the second slave are selectively connected via the 2-port Ethernet communication modules.

Abstract: Provided is a network system and a method for determining a network path, the network system including a plurality of extension units, each containing dualized basic units and dualized extension Ethernet modules, a first ring network connected by a first basic unit and a plurality of first extension Ethernet modules, and a second ring network connected by a second basic unit and a plurality of second extension Ethernet modules, wherein the first extension Ethernet module or the second extension Ethernet module transmits or receives data along a network path determined by judging, by the first extension Ethernet module or the second extension Ethernet module, if the first ring network is connected to the second extension Ethernet module using an intrinsic number of each extension Ethernet module.

Abstract: A method for electing a ring manager of ring topology network and a node are disclosed, where, to this end, a node responsible for electing the ring manager periodically requests each node of diagnostic information on packet traffic and receives the diagnostic information on packet traffic from each node, newly elects a node capable of performing a ring manager role, and informs each node of information on the newly elected ring manager, whereby, the ring network can be more effectively operated because the ring manager is elected based on network load status that changes in real time even in the course of a ring network being initially set up and operated responsive to structural conditions including hop count of each nod and MAC node.

Abstract: The present invention relates to an apparatus and method for monitoring abnormal state of a vehicle. In the method, CAN data collected from an ECU mounted on the vehicle is transformed into coordinates. The coordinates are applied to a distribution map in a specific space, a number of clusters is calculated based on results of the application, and an initial center point corresponding to the number of clusters is selected. Clustering is performed based on the initial center point, and then clusters are generated. At least one piece of data is extracted from each of the clusters, and a state feature of a corresponding cluster is decided on using a difference between maximum and minimum values of attributes constituting the at least one piece of data. A current state of the vehicle is monitored based on current CAN data of the vehicle and state features of the clusters.

Abstract: Disclosed is a hardware protocol stack, where header information of analysis-subjected protocol is stored in a register unit, comparison is made whether information recorded in the header of inputted frame mutually matches header information stored in the register unit, and data is extracted as a result of the comparison.

Abstract: Provided is a network system and a method for determining a network path, the network system including a plurality of extension units, each containing dualized basic units and dualized extension Ethernet modules, a first ring network connected by a first basic unit and a plurality of first extension Ethernet modules, and a second ring network connected by a second basic unit and a plurality of second extension Ethernet modules, wherein the first extension Ethernet module or the second extension Ethernet module transmits or receives data along a network path determined by judging, by the first extension Ethernet module or the second extension Ethernet module, if the first ring network is connected to the second extension Ethernet module using an intrinsic number of each extension Ethernet module.

Abstract: An electrical connector assembly for an electronic device. The electrical connector assembly having an electrical connector and a signal indicator. The electrical connector having at least two communication interfaces utilizing different communication standards. The signal indicator is secured along one side of the electrical connector, wherein the signal indicator emits different colors of light corresponding to a function of one of the at least two communication interfaces.