Abstract: Provided is an image processing device configured to compress first image data. The image processing device includes an encoding circuit configured to compress the first image data into second image data including prediction data and residual data, compress the second image data into third image data by performing entropy encoding on the second image data, generate a header representing a compression ratio of the third image data, and store the third image data along with the header in a memory device as compressed first image data.

Abstract: The present invention relates to a method for improving reliability of artificial intelligence-based object recognition using collective intelligence-based mutual verification, in which: in a server interworking with an artificial intelligence module, one or more object regions included in learning data including a structured image or an unstructured image of a recognition target object to be recognized through the artificial intelligence module are recognized; object region recognition data which sets the recognized object regions is extracted and is provided to one or more user terminals in a designated order; a procedure for receiving, from the user terminals, object region selection data which selects at least one effective object region corresponding to the recognition target object among one or more object regions included in the object region recognition data is performed for a predetermined period of time; and the object region selection data of respective users received from the user terminals is mut

Abstract: The present disclosure relates to a novel concept of sauce, and specifically to a novel sauce using tempeh and a preparation method thereof. A method for preparing a tempeh-containing sauce according to the present application includes secondary fermenting tempeh with brine and other grains to prepare a fermented aged product.

Abstract: Provided is a method and apparatus for generating mixed reality contents. The method includes capturing an object with an advertising board that includes a background part formed of a first color and a line-form advertising part formed of a second color in a background and generating a first image; extracting and removing the background part from the first image; extracting the line-form advertising part; regarding the extracted line-form advertising part as the background part and removing the background part when the second color is identical to a predetermined color; obtaining a second image, in which the background part and the advertising part are removed; and obtaining a third image, which is a background image, and combining the images.

Abstract: Provided is a method of performing an input operation through motion recognition in a glasses-type display device, which is capable of controlling the glasses-type display device through the most natural and intuitive hand motion. The method of performing the input operation through motion recognition in the glasses-type display device includes: an image acquiring operation of acquiring sequential images by using a camera mounted on the glasses-type display device; a hand recognizing operation of recognizing a hand of a user from the acquired sequential images; a hand tracking operation of tracking the recognized hand; a hand motion recognizing operation of recognizing a motion of the hand based on a position of the hand acquired in the hand tracking operation; and a control signal outputting operation of outputting a control signal corresponding to the motion of the hand.

Abstract: Provided is a method and apparatus for extracting an object for a sticker image. The method includes obtaining a first image and a second image by using a first camera, obtaining a third image by using a second camera, determining an object region by using a difference image between the first image and the second image, extracting an object region image that corresponds to the determined object region from the third image, and superimposing by overlaying the extracted object region image with a background image.

Abstract: The present invention relates to a method for enabling motion recognition without recognizing a hand and a remote controller for motion recognition which can suggest an efficient motion and can be implemented at a low cost in a more stable manner. The remote controller for motion recognition, according to the present invention, comprises: an image acquisition unit for acquiring consecutive images of a hand; a difference image configuration unit for obtaining difference images of the acquired consecutive images; a hand trajectory analysis unit for analyzing a moving trajectory of the hand from the difference images; a motion recognition unit for recognizing a hand motion by analyzing the hand trajectory; and a remote signal transmitter for transmitting a remote signal corresponding to the recognized hand motion such that a device to be controlled can be controlled.

Abstract: Provided is a method of performing an input operation through motion recognition in a glasses-type display device, which is capable of controlling the glasses-type display device through the most natural and intuitive hand motion. The method of performing the input operation through motion recognition in the glasses-type display device includes: an image acquiring operation of acquiring sequential images by using a camera mounted on the glasses-type display device; a hand recognizing operation of recognizing a hand of a user from the acquired sequential images; a hand tracking operation of tracking the recognized hand; a hand motion recognizing operation of recognizing a motion of the hand based on a position of the hand acquired in the hand tracking operation; and a control signal outputting operation of outputting a control signal corresponding to the motion of the hand.

Abstract: The present invention relates to a method for enabling motion recognition without recognizing a hand and a remote controller for motion recognition which can suggest an efficient motion and can be implemented at a low cost in a more stable manner. The remote controller for motion recognition, according to the present invention, comprises: an image acquisition unit for acquiring consecutive images of a hand; a difference image configuration unit for obtaining difference images of the acquired consecutive images; a hand trajectory analysis unit for analyzing a moving trajectory of the hand from the difference images; a motion recognition unit for recognizing a hand motion by analyzing the hand trajectory; and a remote signal transmitter for transmitting a remote signal corresponding to the recognized hand motion such that a device to be controlled can be controlled.

Abstract: Provided is a new type of virtual mouse driving method which is independent from individual skin color and capable of being implemented in general environments having a certain degree of disturbance. The virtual mouse driving method according to the invention in which a method of driving the virtual mouse is controlled by a change of hand shape includes an input step of receiving a plurality of images captured by an imaging camera at mutually different time points, a difference image extracting step of extracting a difference image among a plurality of images, and a virtual mouse driving step based on the extracted difference image.

Abstract: An Intermittently Decanted Extended Aeration (IDEA) process characterized in that an aeration period, a settling period and a decanting period are successively repeated in that order in a single reactor, and influent wastewater is transversely supplied at the bottom of the reactor and slowly diffused in the reactor while treated wastewater wich is intermittently discharged during the decanting period is provided. According to the process of the invention, wastewater can be efficiently treated without by-pass of wastewater under dry weather and wet weather conditions as well as stormy weather condition.