| 摘要 |
The present invention relates to a system for predicting crime occurrence in the future through an image analysis server, which comprises a main controller composed of a microprocessor, an image analysis API, an object recognition module, and a transceiver. The image analysis API: receives an original image data recorded by a digital surveillance camera, and stores the original image data in a long-term storage memory; and chain-codes a region, in which a pixel change occurs in the original image data, into a rectangular object, and transmits the digital zoomed-in rectangular object image to the object recognition module. The transceiver measures transmission traffic of a center control server in an integrated security center, connected through a network, and transmits the same to the main controller. The main controller transmits a low capacity image compression file, in which the original image data received by the object recognition module is encryption-compressed into a low image quality and low capacity file of less than or equal to the traffic measured in the transceiver, to the center control server in the integrated security center through a network. The object recognition module transmits object classified images, in which objects in the rectangular object image are classified, and motion pattern image, in which motion patterns of the object images are extracted, to the main controller. The main controller: compares the rectangular object image with the motion patterns and dangerous objects stored in the long-term storage memory; and transmits a dangerous event compression file, which is encryption-compressed by assigning priority feature values thereto, to the center control server in the integrated security center. |
