35results about How to "Save data space" patented technology

The invention provides a high-speed low-power-consumption embedded memory card, comprising a memory for storing data, a main control circuit and a command filtering module, wherein the main control circuit is connected with the memory and used for controlling work of the memory card, and the command filtering module is connected with the main control circuit and a card reader respectively and used for filtering requests of not operating the memory, sent by the card reader. In the invention, the command filtering module is arranged between the card reader and the main control circuit of the memory card and is used for pre-filtering the requests of not carrying out a reading / writing / erasing operation on the memory, so that the working load of the main control circuit can be reduced, the main control circuit is enabled to process the reading / writing / erasing command requests only, the power consumption of the memory card is reduced, the software code compiling is simple, the maintenance is easy, the data space is saved and the response speed of the system is increased. The memory card is especially suitable to being applied to the embedded memory card system having low power consumption and low cost.

The present invention proposes a classification method based on depth feature bags, which extracts the depth features of a representative image of a two-dimensional orthogonal plane, then uses the feature bag model to recode the extracted depth features, and finally uses the kernel fusion method Combining the characteristics of three two-dimensional orthogonal planes, the classification function is obtained to classify the image. This classification method greatly reduces the amount of calculation in the training and testing process, saves data space, and shortens the time of the entire calculation process. The "codebook" is used to encode image features to achieve high-dimensional sparse representation, making features more discriminative. It is more stable and more compact, and can alleviate the overfitting problem caused by insufficient data volume. The three two-dimensional orthogonal planes are multiplied by different weight coefficients when calculating the kernel function, which can make better use of the different spatial information contained in the three different sections, and the constructed classification function is more discriminative.

The invention discloses an unmanned plane borne hyperspectral image detector with a function of synchronization radiation correction, and the detector comprises a target spectrum image detector and an image processor which carries out data communication with the target spectrum image detector. The target spectrum image detector comprises a camera lens, a spectrum splitter and an image detector. One end, opposite to the camera lens, of the spectrum splitter is provided with a signal collection slit. The detector also comprises a reference light collector and a curved surface reflector. The reference light collector is disposed at an upper no-shading place of an unmanned plane. The curved surface reflector is disposed at one end of the signal collection slit of the spectrum splitter. Meanwhile, the invention also provides an unmanned plane borne hyperspectral image detection method with the function of synchronization radiation correction. The detector corrects the offset of the detection data in real time, facilitates the remarkable improvement of the data calculation precision, and is especially suitable for the sweeping detection of an unmanned plane borne hyperspectral image under the varied radiation conditions.

The invention provides a similar data de-duplication method, which includes the following steps of inputting the acquired similar data to a server, extracting the characteristic vectors of the similar data, preprocessing information of the characteristic vectors to acquire the character type index data of the information; performing code conversion to the index data to generate the numerical hashed data of the information; judging whether the hashed data of the characteristic vectors are the same with the standard data stored in a database server or not one by one according to the weight of the information, and then feeding the results back to users. The similar data with different response information can be de-duplicated quickly by the method. Besides, the similar data de-duplication method is high in accuracy and fine in stability.