Abstract: A recognition system for recognizing multiple inputs of gestures, handwriting symbols and virtual keys on a touch screen includes a touch IC serves to convert a plurality of touch signals of the touch screen to a touch data frame. A processor set is connected to the touch IC and serves to perform a touch data processing on the touch data frame. The touch data processing is performed by using a processing directly executed by an OS (Operating System) and a processing of AI (artificial intelligence) recognizing. An AI recognition module is connected to the processor set. The AI recognition module is used for recognizing multiple key inputs, operation gestures and handwriting symbols. The key inputs and handwriting symbols are corrected by a grammar correction and a symbol correction respectively. The touch screen serves to display a virtual keyboard.

Abstract: A handwriting text recognition system based on neural network includes a stroke input processor for receiving tracks of online handwriting texts, a string database for storing a large amount of the tracks; a word recognition neural network; and an after-processor being connected to the string database and the output interface of the text recognition neural network; The handwriting text recognition system based on neural network provides higher rates of confidences. Some natural languages frequently used all over the world can be recognized with a higher accuracy (including languages written from right to left and from left to right). The association relations between the input strokes and the character strings can be provided. It could support any strokes with irregular written orders.

Abstract: A system for determining correction of handwriting Chinese characters includes a Chinese character collector for collecting template Chinese characters which is inputted by handwriting. A feature classifier is connected to a pre-processor for automatically classifying features of the template Chinese characters. Text features of the template Chinese characters are acquired by a program. A tested handwriting Chinese character collector is connected to the feature classifier for collecting the handwriting Chinese characters to be tested. Exact external rectangles and mass centers of the handwriting Chinese characters to be tested are calculated, which are inputted to the feature classifier and the text features for the handwriting Chinese characters to be tested are calculated. A feature comparator is connected to the feature classifier and serves to compare the text features of a handwriting Chinese character to be tested with feature sets stored in a feature set database.

Abstract: A digital trajectory prediction system includes a preprocess unit serving to receive an input trajectory from an input device for preforming a preprocessing on the input trajectory to form a processed trajectory. A neural network prediction model serves to receive the processed trajectory and to output a prediction result including a predicted point. A post-processing unit serves to perform a post-processing on the prediction result for determining whether to add the predicted point to a render point set. A point output unit serves for rendering the render point set. When the predicted point is added to the render point set in the post-processing, the input trajectory and the predicted point are rendered on a screen. When the predicted point is not added to the render point set in the post-processing, only the input trajectory is rendered on the screen.

Abstract: A recognition system for recognizing multiple inputs of gestures, handwriting symbols and virtual keys on a touch screen includes a touch IC serves to convert a plurality of touch signals of the touch screen to a touch data frame. A processor set is connected to the touch IC and serves to perform a touch data processing on the touch data frame. The touch data processing is performed by using a processing directly executed by an OS (Operating System) and a processing of AI (artificial intelligence) recognizing. An AI recognition module is connected to the processor set. The AI recognition module is used for recognizing multiple key inputs, operation gestures and handwriting symbols. The key inputs and handwriting symbols are corrected by a grammar correction and a symbol correction respectively. The touch screen serves to display a virtual keyboard.

Abstract: A handwriting geometry recognition and calibration system by using neural network and mathematical feature includes: a pre-processor for pre-processing coordinate points of geometric figures from user's handwriting so as to get a plurality of sample points which expresses the geometric figures to be recognized; a neural network connected to the pre-processor for receiving the sample points of the geometric figure and recognizing the geometric figure so as to acquire a coarse class of the geometric figure; and an mathematical logic unit connected to the neural network for receiving recognition results from the neural network, including coarse classifications which are used in a secondary classification by using conventional mathematical recognition logics so as to determine an exact geometry shape of the geometric figure; then the geometric figure being calibrated so as to get a geometry with a regular shape.

Abstract: A syntax-directed mathematical expression recognition system combines online handwritten mathematical expression, offline handwritten and printed mathematical expression and structured mathematical expressions to generate a huge amount of mathematical expressions which are usable in training of neural networks. An end to end learning way is used to train the neural network for increment of learning effect and recognition ability. The problem of overfitting in multi-layer neural networks is reduced effectively so as to increase accuracy of testing set. The mode of neural network is usable for online handwritten mathematical expression, offline handwritten and printed mathematical expression. Therefore only one recognition system is used, no other system is needed.

Abstract: A handwriting text recognition apparatus supporting multiple writing directions comprising a processor and a memory; a stroke input processor serving to receive a plurality of online handwriting character string tracks which are exploded into a plurality of strokes; a writing direction determination operator connected to the stroke input processor for calculating a writing direction of the online handwriting character string tracks by using a coordinate of a center of gravity and a plurality of second central moments of a plurality of sampling points of the strokes; a stroke sorting operator connected to the writing direction determination operator and sorting the strokes by the writing direction to obtain an x direction labeled sorting stroke set or a y direction labeled sorting stroke set; and wherein the x direction labeled sorting stroke set or the y direction labeled sorting stroke set is input to a recognition model for recognition.

Abstract: A recognition system for recognizing dual inputs of gestures and virtual keys on a touch screen includes a touch IC connected to the touch screen for receiving a plurality of touch signals generated by the touch screen and converting the touch signals to a touch data frame. A processor set is connected to the touch IC. The processor set serves to perform a touch data processing on the touch data frame from the touch IC. The touch data processing is performed by using a processing directly executed by an OS (Operating System) and a processing of AI (artificial intelligence) recognizing. The processor set serves to perform a grammar correction on a recognition result of the AI recognizing. A result of the grammar correction is output to the OS for processing.

Abstract: A syntax-directed mathematical expression recognition system combines online handwritten mathematical expression, offline handwritten and printed mathematical expression and structured mathematical expressions to generate a huge amount of mathematical expressions which are usable in training of neural networks. An end to end learning way is used to train the neural network for increment of learning effect and recognition ability. The problem of overfitting in multi-layer neural networks is reduced effectively so as to increase accuracy of testing set. The mode of neural network is usable for online handwritten mathematical expression, offline handwritten and printed mathematical expression. Therefore only one recognition system is used, no other system is needed.

Abstract: A projection device includes a front cover, a back cover, a projection lens, a processor, and an electromagnetic touch panel for handwriting input. The electromagnetic touch panel is provided on the inside surface of the front cover and in electrical connection and communication with the processor. When the users writes directly on the front cover or on a writing sheet held by a paper holder on the front cover, the electromagnetic touch panel senses the handwriting, converts it into a writing signal and transmits the writing signal to the projection lens for the handwriting to be displayed on a screen. The projection may further include a capacitive touch panel with a silkscreen printing keypad on one side thereof for the user to key in text. The projection device effectively saves time and space in making presentations or text editing.

Abstract: The dual input device has a keyboard on the front side of the input device, a display screen on the back side of the input device, an electromagnetic touch panel and a control circuit between the front side and the display screen. The keyboard, the display screen, and the electromagnetic touch panel are connected with the control circuit, respectively. The control circuit includes a direction control circuit for controlling the On/Off of the keyboard, the display screen and the electromagnetic touch panel. Through the direction control circuit, the input device combines the traditional keyboard and the writing pad in a single dual input device. In addition, because the electromagnetic touch panel is placed behind the display screen, the input data can be displayed in real time, and the original handwriting can be saved.