Abstract: This invention relates to a method for calculating and adjusting photo and text separation performance. The method for calculating photo and text separation performance comprises a step of generating a standard block attribute file and a step of comparing a photo and text separation result with the standard block attribute file so as to display a photo and text separation performance table. The method for adjusting photo and text separation performance comprises a step of comparing a photo and text separation result with the standard block attribute file to generate the photo and text separation performance table, and a step of according to the photo and text separation performance table to decide whether to use another photo and text separation algorithm or parameters to perform another photo and text separation.

Abstract: An interlocked roll type apparatus comprises: a first spool, two first flanges, a second spool, and two second flanges. The first flange has a first surface, a second surface, and a first periphery surface. The first surface has a first round protruding part. The first periphery surface has at least a first arc part, and the first arc part has a first arc protruding part protruding from the first surface. The second flange has a third surface, a fourth surface, and a second periphery surface. The second periphery surface has at least a second arc part having a second arc protruding part protruding from the third surface. The second arc part and the second arc protruding part are fastened with the first round protruding part of the first flange. The fourth surface has a second arc trench part for fastened with the first arc protruding part of the first flange.

Abstract: A surveillance system including a plurality of network video recorders (NVRs), a plurality of network video capturing devices and a plurality of streaming media presentation devices (SMPDs) is provided. The NVRs can receive and store digitized data packets, as well as transmit video/audio streams, event notifications, authorization codes and control signals. The network video capturing devices, each connected to a corresponding NVR among the plurality of NVRs via a network, can capture video/audio data and providing the digitized data packets by encoding the captured video/audio data. The SMPDs are connected to the plurality of NVRs via the network and can play captured video/audio data and alerting messages associated with the event notifications.

Abstract: A monochromic image processing system and method for background removal according to dynamic data firstly obtain pixel histogram data in part of an image during a photo/text separation process. Afterwards, look-up tables for executing a background removal process are generated immediately to remove background noise of all images simultaneously. This method can solve the problem that in the multi-source document copy mode the background noise of all images cannot be removed at the same time in the prior art. It also increases the flexibility in the background removal process in various copy modes.

Abstract: A printing media loading apparatus includes a roller that has a narrow slit thereon and is used for loading a printing media. There are two nicks respectively on the two sides of the roller, and the narrow slit extends to the two nicks. The printing media loading apparatus further includes at least a detachable fixing unit, installed in a nick on one of the two sides of the roller for fixing the diameter of the roller.

Abstract: A pair of pliers has two bodies, a pair of heads and two connection devices. The bodies are pivotally connected with each other, and each body has a handle and a shank formed respectively at two ends of the body. The heads are detachably attached respectively to the shanks of the bodies. The connection devices are mounted between the heads and the shanks to detachably connect the heads to the shanks. Each connecting device has a connection block and a connection loop. The connecting block is formed on one of the heads and shanks. The connecting loop is formed on a corresponding one of the heads and shanks and securely and is detachably mounted around the connecting block. Accordingly, the pair of pliers is easily changeable for different heads.

Abstract: This specification discloses a system of using a neural network to distinguish text and pictures in an image and the method thereof. Using the knowledge of text recognition learned by the neural network in advance, images data of color brightness and gray levels in an image block are processed to generate a greatest text faith value. The system determines the text status of the image block by comparing a text threshold with the greatest text faith value. If the greatest text faith value is larger than the text threshold, then the image block is determined to contain text pixels; otherwise, the image block contains purely picture pixels. This achieves the goal of separating text and pictures in an image.

Abstract: This invention relates to a method for calculating and adjusting photo and text separation performance. The method for calculating photo and text separation performance comprises a step of generating a standard block attribute file and a step of comparing a photo and text separation result with the standard block attribute file so as to display a photo and text separation performance table. The method for adjusting photo and text separation performance comprises a step of comparing a photo and text separation result with the standard block attribute file to generate the photo and text separation performance table, and a step of according to the photo and text separation performance table to decide whether to use another photo and text separation algorithm or parameters to perform another photo and text separation.

Abstract: A monochromic image processing system and method for background removal according to dynamic data firstly obtain pixel histogram data in part of an image during a photo/text separation process. Afterwards, look-up tables for executing a background removal process are generated immediately to remove background noise of all images simultaneously. This method can solve the problem that in the multi-source document copy mode the background noise of all images cannot be removed at the same time in the prior art. It also increases the flexibility in the background removal process in various copy modes.

Abstract: A photo and text discriminating method is provided. The gray scale image of an input image is obtained, and a plurality of text blocks and a plurality of photo blocks of the gray scale image are distinguished. Whether the text blocks are amplitude-modulation or frequency-modulation halftone photo blocks are determined, and the text blocks are processed by a method different than that used for processing the photo blocks.

Abstract: This specification discloses a system of using a neural network to distinguish text and pictures in an image and the method thereof. Using the knowledge of text recognition learned by the neural network in advance, images data of color brightness and gray levels in an image block are processed to generate a greatest text faith value. The system determines the text status of the image block by comparing a text threshold with the greatest text faith value. If the greatest text faith value is larger than the text threshold, then the image block is determined to contain text pixels; otherwise, the image block contains purely picture pixels. This achieves the goal of separating text and pictures in an image.