Abstract: A method for narrow-band image generation is provided. First obtaining an input image of an object by an image capture unit. Then converting the input image according to an image conversion model and at least one target wave band corresponding to a narrow-band light source to get a simulated narrow-band image. Lastly comparing a simulated narrow-band image information of the simulated narrow-band image with a reference narrow-band image information according to at least one objective similarity index to generate an index data for determining similarity between the simulated narrow-band image information and the reference narrow-band image information. Thereby the simulated narrow-band image is checked by the objective similarity index combined with simulation of narrow-band images using hyperspectral techniques to help doctors in interpretation of endoscopic images.

Abstract: The present application related to a method for deleting redundant images of an endoscope. Firstly, a host receives an input image datum from an endoscopy, hereby, capturing a plurality of corresponded input images from the input image data. Next, a plurality of dimension reduction data are obtained to operate with at least one adjacent dimension reduction datum, and then, a plurality of root mean squared error values are obtained to compare with an error threshold value. Hereby, at least one image unarrived the error threshold value are deleted to obtain a plurality of screened images. Thus, redundant images in the input image data of the endoscopy will be deleted to prevent from performing an image analysis of same images or approximated images.

Abstract: The present invention provides a method for detecting tissue hemorrhage with image analysis. A host produces a plurality of hyperspectral image information according to a plurality of reference images. An image extraction unit extracts an input image to the host. The host transforms the input image according to the plurality of hyperspectral image information to produce a hyperspectral input image. The host produces an input image spectrum according to the hyperspectral input image. The host performs a feature operation on the input image spectrum according to a preset cell band corresponding to a surface cell of small intestine for generating a plurality of corresponding feature bands. The host performs at least one convolution operation on the plurality of feature bands according to a plurality of kernels for producing a convolution result.

Abstract: A lathe protective cover includes a mounting member and a cover unit. The mounting member has a main body including a mounting portion and a pivot portion, is configured to be detachably mounted on and configured to be movable relative to a cutter seat or a workpiece seat of a lathe. The cover unit includes a base plate pivotally connected to the pivot portion and having a through hole, and first and second side plates connected to two opposite sides of and cooperating with the base plate to define a shielding space communicating with the through hole. The cover unit is rotatable relative to the mounting member between a closed position, in which the mounting portion is adjacent to the through hole, and an open position, in which the shielding space and the through hole are away from the mounting portion.

Abstract: A method of performing air pollution estimation is provided. The method is to be implemented using a processor of a computer device and includes: generating a spectral image based on an original color image of an environment under test using a spectral transformation matrix; supplying the spectral image as an input into an estimating model for air pollution estimation; and obtaining an estimation result from the estimating model indicating a degree of air pollution of the environment under test.

Abstract: A method for inspecting authenticity of a hologram is provided. A computer device that stores a color image of the hologram transforms the color image into a hyperspectral image, converts the hyperspectral image into a grayscale image, and determines authenticity of the hologram based on multiple grayscale values in a region of interest in the grayscale image and multiple grayscale thresholds that respectively correspond to different wavelengths.

Abstract: A method of measuring chromaticity of a target object is implemented using a computer device that stores a plurality of light source spectrum datasets each associated with a specific object. The method includes: obtaining a captured color image of the target object; generating a spectral image based on the captured color image using a spectral transformation matrix; obtaining one of the plurality of light source spectrum datasets that is associated with the target object; and calculating a chromaticity dataset of the target object based on the spectral image and the one of the plurality of light source spectrum datasets.

Abstract: A method for measuring a thickness of a thin film layer disposed on a piece of glass is implemented using a computer device that stores a thin film image of the thin film layer, a surface dataset associated with a surface of the thin film layer, and a plurality of reference parameter sets each being associated with a specific thickness of the thin film layer, the method including: generating a spectral image dataset that includes spectral data associated with different pixels of the thin film image using a spectral transformation matrix; performing regression analysis on the surface dataset and the spectral image dataset, so as to obtain a thickness parameter set including a plurality of thickness parameters; and determining a thickness of the thin film layer using the thickness parameter set and the plurality of reference parameter sets.

Abstract: This application provides a method for detecting images of testing object using hyperspectral imaging. Firstly, obtaining a hyperspectral imaging information according to a reference image, hereby, obtaining corresponded hyperspectral image from an input image and obtaining corresponded feature values for operating Principal components analysis to simplify feature values. Then, obtaining feature images by Convolution kernel, and then positioning an image of an object under detected by a default box and a boundary box from the feature image. By Comparing with the esophageal cancer sample image, the image of the object under detected is classifying to an esophageal cancer image or a non-esophageal cancer image. Thus, detecting an input image from the image capturing device by the convolutional neural network to judge if the input image is the esophageal cancer image for helping the doctor to interpret the image of the object under detected.

Abstract: A method for recognizing arteries and veins on a fundus image includes: executing a pre-process operation on the fundus image, so as to obtain a pre-processed fundus image; generating a fundus spectral reflection dataset associated with pixels of the pre-processed fundus image, based on the pre-processed fundus image, and a spectral transformation matrix; obtaining a plurality of principle component scores associated with the pixels of the pre-processed fundus image, respectively; and determining, for each of the pixels of the pre-processed fundus image that has been determined as a part of a blood vessel, whether the pixel belongs to a part of an artery or a part of a vein.

Abstract: A biosensing chip is provided, including a substrate having a photoelectric conversion material, and an electrode disposed on the substrate and including two contact portions and an electrode pattern, wherein the photoelectric conversion material is a monocrystalline silicon material, and the electrode pattern includes micro-electrodes in the form of interdigitated sawtooth. The biosensing chip and the method using the same may distinguish a lesion site of cancer cells and the degree of cancer lesions.

Abstract: The present application discloses a method for detecting image by using semantic segmentation. To input an image with data augmentation, and then encode and decode using a neural network. At least one semantically divided, and finally the at least one semantically divided is compared with the sample to classify as a target or a non-target. In this way, the CNN is used to detect whether the image is the SCC image or not, and locate the section, thereby assisting the doctor in interpreting the esophagus image.

Abstract: The present application related to a method for detecting image using hyperspectral imaging by band selection. Firstly, obtaining a hyperspectral imaging information according to a reference image, hereby, obtaining corresponded hyperspectral image from an input image, and obtaining corresponded feature values by band selection for operating Principal components analysis to simplify feature values. Then, obtaining feature images by Convolution kernel, and then positioning an image of an object under detected by a default box and a boundary box from the feature image. By Comparing with the esophageal cancer sample image, the image of the object under detected is classifying to an esophageal cancer image or a non-esophageal cancer image. Thus, detecting an input image from the image capturing device by the convolutional neural network to judge if the input image is the esophageal cancer image for helping the doctor to interpret the image of the object under detected.

Abstract: A method for detecting objects in hyperspectral imaging is revealed. First obtaining a hyperspectral imaging information by a reference image. Then converting an input image according to the hyperspectral imaging information to get a hyperspectral image. A plurality of hyperspectral eigenvalues is obtained after image analysis of the hyperspectral image. Next getting a plurality of dimensionality reduction eigenvalues by a principal component analysis (PCA). Then performing convolution operation on the dimensionality reduction eigenvalues to get a value of a convolution matrix for extracting a feature image from an image of an object to be detected in the input image. Generating an anchor box and a prediction box in the feature image to get a bounded image. Lastly matching and comparing the bounded image with a sample image to determine whether the input image is a target object image. Thereby the method provides assistance for physicians in gastrointestinal image diagnosis.

Abstract: A method of measuring chromaticity of a target object is implemented using a computer device that stores a plurality of light source spectrum datasets each associated with a specific object. The method includes: obtaining a captured color image of the target object; generating a spectral image based on the captured color image using a spectral transformation matrix; obtaining one of the plurality of light source spectrum datasets that is associated with the target object; and calculating a chromaticity dataset of the target object based on the spectral image and the one of the plurality of light source spectrum datasets.

Abstract: The present application related to a method for detecting an object image using a convolutional neural network. Firstly, obtaining feature images by Convolution kernel, and then positioning an image of an object under detected by a default box and a boundary box from the feature image. By Comparing with the sample image, the detected object image is classifying to an esophageal cancer image or a non-esophageal cancer image. Thus, detecting an input image from the image capturing device by the convolutional neural network to judge if the input image is the esophageal cancer image for helping the doctor to interpret the detected object image.

Abstract: This application provides a method for detecting images of testing object using hyperspectral imaging. Firstly, obtaining a hyperspectral imaging information according to a reference image, hereby, obtaining corresponded hyperspectral image from an input image and obtaining corresponded feature values for operating Principal components analysis to simplify feature values. Then, obtaining feature images by Convolution kernel, and then positioning an image of an object under detected by a default box and a boundary box from the feature image. By Comparing with the esophageal cancer sample image, the image of the object under detected is classifying to an esophageal cancer image or a non-esophageal cancer image. Thus, detecting an input image from the image capturing device by the convolutional neural network to judge if the input image is the esophageal cancer image for helping the doctor to interpret the image of the object under detected.

Abstract: A method for analyzing 2D material thin film and a system for analyzing 2D material thin film are disclosed. The detection method includes the following steps: capturing sample images of 2D material thin films; measuring the 2D material thin films by a Raman spectrometer; performing a visible light hyperspectral algorithm on the sample images by a processor to generate a plurality of visible light hyperspectral images; performing a training and validation procedure, performing an image feature algorithm on the visible light hyperspectral images, and establishing a thin film prediction model based on a validation; and capturing a thin-film image to be measured by the optical microscope, performing the visible light hyperspectral algorithm, and then generating a distribution result of the thin-film image to be measured according to an analysis of the thin film prediction model.

Abstract: A method of performing air pollution estimation is provided. The method is to be implemented using a processor of a computer device and includes: generating a spectral image based on an original color image of an environment under test using a spectral transformation matrix; supplying the spectral image as an input into an estimating model for air pollution estimation; and obtaining an estimation result from the estimating model indicating a degree of air pollution of the environment under test.

Abstract: The present application related to a method for detecting an object image using a convolutional neural network. Firstly, obtaining feature images by Convolution kernel, and then positioning an image of an object under detected by a default box and a boundary box from the feature image. By Comparing with the sample image, the detected object image is classifying to an esophageal cancer image or a non-esophageal cancer image. Thus, detecting an input image from the image capturing device by the convolutional neural network to judge if the input image is the esophageal cancer image for helping the doctor to interpret the detected object image.