Abstract: This application provides a two-dimensional code-based data processing method, apparatus, and system, where the two-dimensional code-based data processing method includes: receiving and parsing a material to be parsed to obtain a two-dimensional code and a target material, where the two-dimensional code carries target label information; recognizing the two-dimensional code to obtain the target label information; and storing the target material into a material library based on the target label information. The target material is quickly labeled based on the two-dimensional code, so that the target material is automatically imported into the corresponding material library, to help a user subsequently search for and manage the target material.

Abstract: A deproteinized natural rubber latex and preparation method thereof are provided. A combined immobilization technology of a multi-enzyme system is introduced to form a multi-enzyme synergistic effect, such that high-concentration enzymes contact with proteins in natural rubber latex. Meanwhile, due to the complexity and diversity of latex proteins, the catalytic properties of different enzymes are effectively combined to realize a multi-enzyme cascade reaction, and the proteins are hydrolyzed gradually and orderly. In addition, an immobilized enzyme filter screen is introduced, and the filter screen is lifted and circulated to make rubber particles collide with high-density enzyme carrier and then complete mixing evenly to reduce damages to the rubber particles when surface proteins are removed, thereby maintaining original physical properties of the latex. The rubber latex has high stability, reusability, is easy for separation and continuous operation, and is environmentally friendly.

Abstract: A system to create a stacked classifier model combination or classifier ensemble has been designed for identification of undisclosed flaws in software components on a large-scale. This classifier ensemble is capable of at least a 54.55% improvement in precision. The system uses a K-folding cross validation algorithm to partition a sample dataset and then train and test a set of N classifiers with the dataset folds. At each test iteration, trained models of the set of classifiers generate probabilities that a sample has a flaw, resulting in a set of N probabilities or predictions for each sample in the test data. With a sample size of S, the system passes the S sets of N predictions to a logistic regressor along with “ground truth” for the sample dataset to train a logistic regression model. The trained classifiers and the logistic regression model are stored as the classifier ensemble.

Abstract: A system to create a stacked classifier model combination or classifier ensemble has been designed for identification of undisclosed flaws in software components on a large-scale. This classifier ensemble is capable of at least a 54.55% improvement in precision. The system uses a K-folding cross validation algorithm to partition a sample dataset and then train and test a set of N classifiers with the dataset folds. At each test iteration, trained models of the set of classifiers generate probabilities that a sample has a flaw, resulting in a set of N probabilities or predictions for each sample in the test data. With a sample size of S, the system passes the S sets of N predictions to a logistic regressor along with “ground truth” for the sample dataset to train a logistic regression model. The trained classifiers and the logistic regression model are stored as the classifier ensemble.

Abstract: A system to create a stacked classifier model combination or classifier ensemble has been designed for identification of undisclosed flaws in software components on a large-scale. This classifier ensemble is capable of at least a 54.55% improvement in precision. The system uses a K-folding cross validation algorithm to partition a sample dataset and then train and test a set of N classifiers with the dataset folds. At each test iteration, trained models of the set of classifiers generate probabilities that a sample has a flaw, resulting in a set of N probabilities or predictions for each sample in the test data. With a sample size of S, the system passes the S sets of N predictions to a logistic regressor along with “ground truth” for the sample dataset to train a logistic regression model. The trained classifiers and the logistic regression model are stored as the classifier ensemble.

Abstract: A system to create a stacked classifier model combination or classifier ensemble has been designed for identification of undisclosed flaws in software components on a large-scale. This classifier ensemble is capable of at least a 54.55% improvement in precision. The system uses a K-folding cross validation algorithm to partition a sample dataset and then train and test a set of N classifiers with the dataset folds. At each test iteration, trained models of the set of classifiers generate probabilities that a sample has a flaw, resulting in a set of N probabilities or predictions for each sample in the test data. With a sample size of S, the system passes the S sets of N predictions to a logistic regressor along with “ground truth” for the sample dataset to train a logistic regression model. The trained classifiers and the logistic regression model are stored as the classifier ensemble.

Abstract: This invention discloses a UAV inspection method for power line based on human visual system. Image preprocessing module preprocesses the power line image of the input system. Power line detection module uses human visual attention mechanism to complete segmentation of the power line in the image. Binocular image registration module uses SURF algorithm to provide exact match of the feature points. The obstacle detection and early warning module uses binocular visual principle to calculate the three-dimensional coordinates of the matching point and the power line. The result output and feedback module calculates the vertical distance from the matching point to the power line according to the information about the space coordinates to complete feedback of the information about the obstacle with a threat to the power line. The method can accurately analyze the obstacle of power line in a quantitative manner, and the analysis result is stable and objective.

Abstract: This invention discloses a UAV inspection method for power line based on human visual system. Image preprocessing module preprocesses the power line image of the input system. Power line detection module uses human visual attention mechanism to complete segmentation of the power line in the image. Binocular image registration module uses SURF algorithm to provide exact match of the feature points. The obstacle detection and early warning module uses binocular visual principle to calculate the three-dimensional coordinates of the matching point and the power line. The result output and feedback module calculates the vertical distance from the matching point to the power line according to the information about the space coordinates to complete feedback of the information about the obstacle with a threat to the power line. The method can accurately analyze the obstacle of power line in a quantitative manner, and the analysis result is stable and objective.