Abstract: A method for acquiring shadow-free images of a document for scanning or other purposes is applied in a device. The method includes training a shadow prediction model based on sample documents of a sample library and inputting a background color and a shadow mask of each of the sample documents extracted by the shadow prediction model into a predetermined shadow removing network for training, to obtain a shadow removing model. The method further includes obtaining a background color and a shadow mask of the document through the shadow prediction model and removing the shadows of the document based on the shadow removing model. The device utilizing the method is also disclosed.

Abstract: The invention discloses a method of isolating nucleic acid and the system thereof, which relates to the following process: (1) providing a biological sample containing cells; (2) performing a cell concentration procedure; (3) performing a magnetic separation procedure through a magnetic separation unit; (4) performing a suspension procedure through a suspending unit; (5) performing a lysis procedure through a lysis unit; and (6) performing a nucleic acid extraction procedure to extract the nucleic acids from the biological samples; wherein the cell concentration procedure does not involve any means of centrifugation. Therefore, the method is simple and efficient, and it is particularly favorable in full automation of nucleic acid isolation. Furthermore, the method and the system as a whole further include a clarification procedure executed by a clarification unit without using centrifugation means, so that improves the quality of nucleic acid isolation.

Abstract: A method for recognition of characters by optical means in an unclear or non-optimal image of an object document, the image carrying shadows or other impediments inputs the document into a shadow prediction model to obtain a shadow mask. A determination is made as to whether the shadow mask of the document affect an optical character recognition (OCR) performance. The method further inputs the document into a shadow removing model for removal of shadows to obtain an intermediate document if the shadow mask are deemed to affect the OCR performance, then OCR can then be performed on the final object document.

Abstract: A driving warning method applied to a vehicle-mounted device is provided. The method includes detecting a moving object and a moving direction of the moving object by using at least one external sensor when a vehicle is moving. A driving behavior of a driver of the vehicle is monitored by using at least one internal sensor when the moving object and the moving direction of the moving object are detected. Once the moving direction of the moving object is not parallel to a moving direction of the vehicle and a sight direction of the driver does not cross the moving direction of the moving object, a first warning is transmitted.

Abstract: A method for generating images with high dynamic range (HDR) based on multiple images captured at different aperture values, under different conditions, or at different shutter speeds is applied in a device. The method inputs the original multiple images into a predetermined model and aligns the multiple images. The method further confirms object images that need to be attended among multiple aligned images and obtains a merge weighting for each of the object images, and merges the images for a generated HDR according to the merge weighting of each image. The device utilizing the method is also disclosed.

Abstract: A method for recognition of characters by optical means in an unclear or non-optimal image of an object document, the image carrying shadows or other impediments inputs the document into a shadow prediction model to obtain a shadow mask. A determination is made as to whether the shadow mask of the document affect an optical character recognition (OCR) performance. The method further inputs the document into a shadow removing model for removal of shadows to obtain an intermediate document if the shadow mask are deemed to affect the OCR performance, then OCR can then be performed on the final object document.

Abstract: A method for acquiring shadow-free images of a document for scanning or other purposes is applied in a device. The method includes training a shadow prediction model based on sample documents of a sample library and inputting a background color and a shadow mask of each of the sample documents extracted by the shadow prediction model into a predetermined shadow removing network for training, to obtain a shadow removing model. The method further includes obtaining a background color and a shadow mask of the document through the shadow prediction model and removing the shadows of the document based on the shadow removing model. The device utilizing the method is also disclosed.

Abstract: The invention discloses a method of isolating nucleic acid and the system thereof, which relates to the following process: (1) providing a biological sample containing cells; (2) performing a cell concentration procedure; (3) performing a magnetic separation procedure through a magnetic separation unit; (4) performing a suspension procedure through a suspending unit; (5) performing a lysis procedure through a lysis unit; and (6) performing a nucleic acid extraction procedure to extract the nucleic acids from the biological samples; wherein the cell concentration procedure does not involve any means of centrifugation. Therefore, the method is simple and efficient, and it is particularly favorable in full automation of nucleic acid isolation. Furthermore, the method and the system as a whole further include a clarification procedure executed by a clarification unit without using centrifugation means, so that improves the quality of nucleic acid isolation.

Abstract: A method of training an object recognition model includes obtaining a sample set. The sample set is divided into a training set and a verification set. The object recognition model is obtained by training a neural network using the training set, and the object recognition model is verified using the verification set.

Abstract: A driving warning method applied to a vehicle-mounted device is provided. The method includes detecting a moving object and a moving direction of the moving object by using at least one external sensor when a vehicle is moving. A driving behavior of a driver of the vehicle is monitored by using at least one internal sensor when the moving object and the moving direction of the moving object are detected. Once the moving direction of the moving object is not parallel to a moving direction of the vehicle and a sight direction of the driver does not cross the moving direction of the moving object, a first warning is transmitted.

Abstract: A human-computer interaction method applied to a vehicle-mounted device is provided. The method includes obtaining video data of a scene inside a vehicle which is captured by a camera in real time. An action of a passenger in each of a plurality of seating positions in the vehicle is detected from the video data. Once a specified action is detected, a corresponding control operation is executed based on the specified action and the seating position of the passenger who performs the specified action.

Abstract: A method for removing haze from an image is provided. The image is converted to S(x) and V(x), the S(X) is saturation component of the image, and the V(x) is value component of the image. The S(x) is filtered to obtain S?(x), and the V(x) is filtered to obtain V?(x). A depth d(x) is estimated by calculating a difference between the S?(x) and V?(x). Once a transmission map t(x) is calculated based on the d(x), haze is removed from the image based on the transmission map t(x).

Abstract: A method of creating and sharing a streaming video to other devices by an electronic device, which comprises a first camera and a second camera, is disclosed. A voice signal collected by the electronic device is encoded into digital audio data. The first camera captures first still images and the second camera captures second still images simultaneously. The first still images and/or the second still images are cropped and combined to generate combined images. The combined images are encoded into encoded images with a predetermined format. A streaming video is obtained by mixing the encoded images with the digital audio data and transmitted to other remote devices.

Abstract: A method of creating and sharing a streaming video to other devices by an electronic device, which comprises a first camera and a second camera, is disclosed. A voice signal collected by the electronic device is encoded into digital audio data. The first camera captures first still images and the second camera captures second still images simultaneously. The first still images and/or the second still images are cropped and combined to generate combined images. The combined images are encoded into encoded images with a predetermined format. A streaming video is obtained by mixing the encoded images with the digital audio data and transmitted to other remote devices.

Abstract: A method for removing haze from an image is provided. The image is converted to S(x) and V(x), the S(X) is saturation component of the image, and the V(x) is value component of the image. The S(x) is filtered to obtain S?(x), and the V(x) is filtered to obtain V?(x). A depth d(x) is estimated by calculating a difference between the S?(x) and V?(x). Once a transmission map t(x) is calculated based on the d(x), haze is removed from the image based on the transmission map t(x).

Abstract: A method of creating and sharing a streaming video to other devices by an electronic device, which comprises a first camera and a second camera, is disclosed. A voice signal collected by the electronic device is encoded into digital audio data. The first camera captures first still images and the second camera captures second still images simultaneously. The first still images and/or the second still images are cropped and combined to generate combined images. The combined images are encoded into encoded images with a predetermined format. A streaming video is obtained by mixing the encoded images with the digital audio data and transmitted to other remote devices.

Abstract: Methods are disclosed for isolating nucleic acids from formalin-fixed paraffin embedded (FFPE) tissue samples. Each of tissue samples contains paraffin and a target biological tissue or material, and the method includes the steps of: adding a first reagent and a second reagent to the FFPE tissue sample, the first reagent dissolving the paraffin material and the second reagent lysing the biological tissue; mixing the first reagent, the second reagent, and the FFPE tissue sample to form a first mixture; (2) heating the first mixture at 50-80° C. for 30-90 minutes; and then heating the first mixture at 80-95° C. for 30-90 minutes to fractionize the first mixture to form an aqueous phase and an oil phase; (3) collecting an aqueous solution from the aqueous phase; and (4) isolating nucleic acids from the aqueous solution. The method improves the efficiency and convenience of isolating nucleic acids from FFPE tissue samples.

Abstract: Methods are disclosed for isolating nucleic acids from formalin-fixed paraffin embedded (FFPE) tissue samples. Each of tissue samples contains paraffin and a target biological tissue or material, and the method includes the steps of: adding a first reagent and a second reagent to the FFPE tissue sample, the first reagent dissolving the paraffin material and the second reagent lysing the biological tissue; mixing the first reagent, the second reagent, and the FFPE tissue sample to form a first mixture; (2) heating the first mixture at 50-80° C. for 30-90 minutes; and then heating the first mixture at 80-95° C. for 30-90 minutes to fractionize the first mixture to form an aqueous phase and an oil phase; (3) collecting an aqueous solution from the aqueous phase; and (4) isolating nucleic acids from the aqueous solution. The method improves the efficiency and convenience of isolating nucleic acids from FFPE tissue samples.

Abstract: Magnetic polymer microbeads and a method for preparing the same are provided. The method for preparing the magnetic polymer microbeads includes the following steps: preparing polymer particles; immersing the polymer particles into a solution in order to swell the polymer particles; adding magnetic nanoparticles to the solution and allowing the magnetic nanoparticles to enter an interior of the polymer particles; and separating the polymer particles from the solution, wherein the polymer particle is made of polystyrene, or a copolymer containing styrene, and the solution includes a medium polar solvent. The average particle size of the magnetic polymer microbeads of the present invention ranges from submicrons to microns. The magnetic polymer microbeads of the present invention have high magnetization, and the various functional groups can be introduced onto the surfaces thereof.

Abstract: Magnetic polymer microbeads and a method for preparing the same are provided. The method for preparing the magnetic polymer microbeads includes the following steps: preparing polymer particles; immersing the polymer particles into a solution in order to swell the polymer particles; adding magnetic nanoparticles to the solution and allowing the magnetic nanoparticlse to enter an interior of the polymer particles; and separating the polymer particles from the solution, wherein the polymer particle is made of polystyrene, or a copolymer containing styrene, and the solution includes a medium polar solvent. The average particle size of the magnetic polymer microbeads of the present invention ranges from submicrons to microns. The magnetic polymer microbeads of the present invention have high magnetization, and the various functional groups can be introduced onto the surfaces thereof.