Abstract: A control method of automatic driving imported “Smart Gains” model can bypass the problem caused by high levels of complexity that currently trouble automatic driving control systems. The knowledge generated by a Gaussian process machine learning model with the maximum probability can carry out the closed-loop control of automatic driving with a given trajectory, can solve the nonlinear adjustment problem of the actuators of the automatic driving vehicle, as well as the optimization control problem of the randomness of the control object. This feature can also make the automatic driving vehicle run smoothly, save energy, be comfortable and fast, and achieve automatic driving above Level 4.

Abstract: The invention proposes an automatic driving “machine consciousness” model, which is composed by the human's safety driving rules. Establish the dynamic fuzzy event probability measure relation, or fuzzy relation, or probability relation of the automatic driving vehicle and the surrounding passing vehicle. The decision result of “machine consciousness” of automatic driving vehicle is realized by complicated logic operation and using the antagonistic result of logic operation in both positive and negative directions. The implementation result is that it can make the decision-making result of automatic driving vehicle close to the result of human's biological consciousness, which can improve the safety of automatic driving vehicle, reduce the development cost and reduce the distance of road test.

Abstract: The invention refers to an image extraction method importing into SDL model in the field of information processing, and is characterized in that the target image are labeled artificially for plural times on computer data. The target image that has been labeled to be extracted for plural times will obtain the maximum probability value and scale of each parameter constituting the target image by machine learning. And the target image can be obtained from the sample computer data according to the maximum probability value or its scale range. The implementation effect of this method is to extract the required image arbitrarily from an image, eliminate the interference of background image affecting the result of image recognition, and improve the effect of image processing and the accuracy of image recognition, which is a new image processing algorithm that subverts the traditional binaryzation algorithm.

Abstract: A method for simulating a deep learning model of function mapping uses algorithms that can be calculated numerically. In a functional mapping model of simulated deep learning by an algorithm, a SDL model enables fusion with a Gaussian distribution model. By combining two Gaussian distribution models and the mapping of functions, both features can be exhibited, and a powerful artificial intelligence model can be constructed. The SDL model clustering algorithm is the fusion of the function mapping model and the Gaussian distribution model. Optimal clustering of feature vectors is done through probability scale self-organization and probability space distances. The simulation method does not need a combination method as in conventional deep learning to obtain the training data to be identified. Thus, the support of big hardware such as GPU-like deep learning is not needed, black box problems do not occur, and there is no need for enormous data annotation work.

Abstract: A method for simulating a deep learning model of function mapping uses algorithms that can be calculated numerically. In a functional mapping model of simulated deep learning by an algorithm, a SDL model enables fusion with a Gaussian distribution model. By combining two Gaussian distribution models and the mapping of functions, both features can be exhibited, and a powerful artificial intelligence model can be constructed. The SDL model clustering algorithm is the fusion of the function mapping model and the Gaussian distribution model. The simulation method does not need a combination method as in conventional deep learning to obtain the training data to be identified. Thus, the support of big hardware such as GPU-like deep learning is not needed, black box problems do not occur, and there is no need for enormous data annotation work. Using small amount of training data can get the results of large data set training and achieve lower costs.

Abstract: In the current artificial intelligence field, models of deep learning that is prevalent can only map functions. Therefore, a machine learning model with higher performance is desirable. The issue is to construct a machine learning model that enables deep competitive learning between data based on the exact distance. A precise distance scale is submitted by unifying Euclidean space and probability space. It submits a measure of the probability measure of fuzzy event based on this distance. Or, it constructs a new neural network that can transmit information of the maximum probability. Furthermore, super deep competition learning is performed between data having very small ambiguous fuzzy information and minute unstable probability information. By performing integral calculation on this result, it has become possible to obtain dramatic effects at the macro level.

Abstract: In the current artificial intelligence field, models of deep learning that is prevalent can only map functions. Therefore, a machine learning model with higher performance is desirable. The issue is to construct a machine learning model that enables deep competitive learning between data based on the exact distance. A precise distance scale is submitted by unifying Euclidean space and probability space. It submits a measure of the probability measure of fuzzy event based on this distance. Or, it constructs a new neural network that can transmit information of the maximum probability. Furthermore, super deep competition learning is performed between data having very small ambiguous fuzzy information and minute unstable probability information. By performing integral calculation on this result, it has become possible to obtain dramatic effects at tape macro level.

Abstract: The invention proposes an automatic driving “machine consciousness” model, which is composed by the human's safety driving rules. Establish the dynamic fuzzy event probability measure relation, or fuzzy relation, or probability relation of the automatic driving vehicle and the surrounding passing vehicle. The decision result of “machine consciousness” of automatic driving vehicle is realized by complicated logic operation and using the antagonistic result of logic operation in both positive and negative directions. The implementation result is that it can make the decision-making result of automatic driving vehicle close to the result of human's biological consciousness, which can improve the safety of automatic driving vehicle, reduce the development cost and reduce the distance of road test.

Abstract: The invention refers to an image extraction method importing into SDL model in the field of information processing, and is characterized in that the target image are labeled artificially for plural times on computer data. The target image that has been labeled to be extracted for plural times will obtain the maximum probability value and scale of each parameter constituting the target image by machine learning. And the target image can be obtained from the sample computer data according to the maximum probability value or its scale range. The implementation effect of this method is to extract the required image arbitrarily from an image, eliminate the interference of background image affecting the result of image recognition, and improve the effect of image processing and the accuracy of image recognition, which is a new image processing algorithm that subverts the traditional binaryzation algorithm.

Abstract: The invention proposes a control method of automatic driving imported “Smart Gains” model, which is characterized by that it can bypass the NP problem caused by the high complexity that currently troubles the automatic driving control, and though the prior knowledge generated by the Gaussian process machine learning model with the maximum probability, it can carry out the closed-loop control of automatic driving with a given trajectory, which can solve the nonlinear adjustment problem of the actuators of the automatic driving vehicle, and the optimization control problem of the randomness of the control object, and can make the automatic driving vehicle run smoothly, save energy, be comfortable and fast, and desire to get the level of automatic driving above L4.

Abstract: In the current artificial intelligence field, models of deep learning that is prevalent can only map functions. Therefore, a machine learning model with higher performance is desirable. The issue is to construct a machine learning model that enables deep competitive learning between data based on the exact distance. A precise distance scale is submitted by unifying Euclidean space and probability space. It submits a measure of the probability measure of fuzzy event based on this distance. Or, it constructs a new neural network that can transmit information of the maximum probability. Furthermore, super deep competition learning is performed between data having very small ambiguous fuzzy information and minute unstable probability information. By performing integral calculation on this result, it has become possible to obtain dramatic effects at tape macro level.

Abstract: A method for constructing an artificial intelligence super depth learning model includes inputting by an objective function to input into each node of an input layer by interposing a no-teacher machine learning, connecting the no-teacher machine learning mutually between each node of the input layer and a nerve layer, calculating output reference values based on a learning value obtained by the no-teacher machine learning, a trigger threshold of a cranial nerve, or a sampling learning value, and determining an excitation level according to output reference values of all the nerve layers.

Abstract: A code conversion device for image information for generating an image code which is unique to the image information from the image information, the code conversion device for image information includes a processor and a memory, wherein the memory contains instructions for causing the processor to perform operations of: converting acquired raw image information into a plurality of pieces of developed image information; extracting each piece of feature information, from each of the plurality of pieces of developed image information by performing a self-organization processing using a probability scale on each of the plurality of pieces of developed image information; and quantifying a plurality of pieces of feature information and generating an image code.

Abstract: In the field of information processing, the invention is related to an information embedded code, a method for generating an information embedded code, a method for embedding the code, and a method for reading the same. The invention is characterized by enabling distinguishing of product authenticity under natural light and using a standard mobile phone, and solves the problem of unifying authenticity identification for a general consumer and authenticity identification of an expert. An information embedded code is configured from a dot pattern enabling multi bits information to be written on the basis of the geometric placement or physical placement of an information dot relative to a specified reference dot. By adapting a new type of vertically-horizontally integrated virtual reference line the information embedded code according to this invention is capable of increasing the efficiency of code information recording.

Abstract: An image code conversion device for image information and its method are provided. By extracting a plurality of features of an image based on raw image information taken by an image reading device and converting the features into a unique code, the image code conversion device for image information and its method can stably convert the same image into the same image code. Raw image information acquired by an image reading device is converted into a plurality of pieces of developed image information on the basis of geometrical or physical factors. From the plurality of developed image information, feature information is quantified by a self-organization processing based on a probability scale. The quantified feature information is converted into a digital code. Accordingly, a unique image code corresponding to the raw image information is generated.

Abstract: The present invention provides a method of generating information embedded halftone screen code. According to this method, massive digital information can be stored through printing on at least one type of print media. The information embedded can be read and recognized simply and reliably. The quality of the images will not be reduced after information embedded. The information printed on the media includes a predetermined array of halftone dots with different morphology including physical and geometrical characteristics, which forms the computer codes to embed information into printed content. The advantages of this invention are: the maximum similarity value can be reduced to under the threshold value according to this method; the recognition performance of the halftone screen code can be improved; and robustness can still be maintained at a high level even though the paper is defected or polluted.

Abstract: The present invention provides a method of generating information embedded halftone screen code. According to this method, massive digital information can be stored through printing on at least one type of print media, such as images, texts, symbols, background, and so on. And the information embedded can be read and recognized simply and reliably. Particularly, the quality of the images will not be reduced after information embedded. The information printed on the media comprises a predetermined array of halftone dots with different morphology including physical and geometrical characteristics, which forms the computer codes to embed information into printed content. The advantages of this invention are: the maximum similarity value can be reduced to under the threshold value according to this method; the recognition performance of the halftone screen code can be improved; and robustness can still be maintained at a high level even though the paper is defected or polluted.