35 results about "Narrow-band imaging" patented technology

The invention discloses a device and a method for cooperatively detecting a moving target by using all-optical-waveband (including ultraviolet, visible, near-infrared, medium-wave infrared and long-wave infrared) maps. The device comprises a large field-of-view two-dimensional scanning sighting telescope, a common aperture primary optical system module, an infrared imaging and spectrum forming optical subsystem module, an ultraviolet/visible/near-infrared spectrum forming and visible near-infrared imaging optical subsystem module, a short/medium/long-wave infrared spectrum measuring module, a medium-wave wide/narrow band imaging module, a visible near-infrared spectrum measuring module, a visible near-infrared imaging module, an ultraviolet measuring module, a map fusion signal processing module, a control module and a servo system. The device and the method utilizes medium-wave infrared imaging and visible near-infrared imaging for recognizing a suspected moving target and guides spectrum measurement, completes the final recognition of the suspected target with cooperation of spectrum measurement data, and solves the difficulties of the existing detection device such as incomplete detection bands, limited optical path layout, large equipment size, few types of detected moving targets and dynamic changing objects, and poor detection capability.

A light source device (20) comprises a first rotational filter part (23) in which a blue filter (23a), a green filter (23b), and a magenta filter (23c) can be disposed in a light path (LP), a second rotational filter part (24) in which an yellow filter (24a) can be disposed therein, and a band selection filter part (22) in which an NBI filter for limiting blue or green light to a narrow band can be disposed therein. The first and second rotational filter parts (23, 24) can be controlled so that in the case of normal light imaging, the yellow filter (24a) is disposed in the light path (LP) when the magenta filter (23c) is disposed in the light path (LP), and in the case of narrow band imaging, the NBI filter is disposed in the light path (LP), and the yellow filter (24a) is disposed in thelight path (LP) when the green filter (23b) is disposed in the light path (LP).

A smooth procession cone parameter estimation method based on high-resolution ISAR imaging mainly solves the problems that joint estimation on cone geometrical parameters and motion parameters in the parameter estimation process of a smooth cone is difficult to carry out, and the calculated amount is large and the needed prior information amount is large in narrow-band imaging and parameter estimation. The method includes the steps of 1, obtaining frequency in ISAR recording, namely, a slowness-time domain echo, 2, estimating the number of Doppler ambiguity times of the echo by utilizing a curve fitting method, 3, compensating the number of Doppler ambiguity times of the echo, 4, correcting range walk of the echo, 5, carrying out the ISAR imaging on echo data by utilizing a matched filtering method, and accurately estimating position information of a scattering center, and 6, carrying out the joint estimation on cone sizes, radar sights and angles of precession by utilizing a least square fitting method. The smooth procession cone parameter estimation method has the advantages of being easy to operate, high in estimation accuracy, small in amount of needed prior information, and capable of carrying out the joint estimation on the cone geometrical parameters and the motion parameters.

The invention relates to the technical field of medical technology assistance, in particular to an artificial intelligence diagnosis method for gastric cancer under a narrow-band imaging amplificationgastroscope, which comprises the following steps: S1, constructing a mini-UNet neural network model; S2, constructing a UNet + + image segmentation neural network model, and obtaining an area ratio Rabnormal of an image feature difference region; S3, adopting a generative adversarial network (GAN) technology to obtain a microvascular morphological diagram and a microstructure morphological diagram of the feature abnormal region; S4, identifying the microvessel form dissimilarity degree and the microstructure form dissimilarity degree in the microvessel form diagram and the microstructure formdiagram by the neural network model ResNet50; and S5, carrying out identification and judgment by using the random forest model obtained by training to obtain final judgment of cancer or non-cancer,and identifying the canceration position range of the image, which is judged to be cancer, as the identified image feature difference region Pabnormal. The sensitivity and specificity of the kit to cancer and non-cancer recognition reach about 93.4% and 90.7% respectively, a clinician can be effectively assisted in discriminating and diagnosing cancer and non-cancer, and a canceration position range is given.

The invention discloses a narrow-band imaging endoscope device which comprises a light source structure, an endoscope structure, an image processing module, a synchronous irradiating assembly and a spectrum light splitting assembly. The light source structure irradiates at least one beam of illuminating light provided with regulated wavelength frequency bands on detected tissue. The endoscope structure photographs an image of the detected tissue irradiated by the illuminating light provided with the regulated wavelength frequency bands. The image processing module processes the image photographed by the endoscope structure. The synchronous irradiating assembly irradiates the illuminating light emitted by the light source structure on the surface of the detected tissue synchronously. The spectrum light splitting assembly splits the light reflected back by the detected tissue and then inputs the light into the image processing module. The narrow-band imaging endoscope device effectively solves the problem that due to motion of the detected tissue, a narrow-band image is distorted, the structure of the endoscope device is simplified, the reliability of the endoscope device is improved, and consumed energy is reduced.

The invention discloses an optical system for infrared medium and long wave spectrum imaging. The optical system comprises a board band lens assembly and a multi-spectral filter assembly. The board band lens assembly is arranged in front of the multi-spectral filter assembly and used for optical wave focus imaging. The multi-spectral filter assembly is used for conducting filtering and cofocal plane compensation on optical wave used for focusing the broad band lens assembly. The optical system for imaging is large in hole diameter which can reach 0.9-1.1, and imaging quality approaches diffraction limit. The optical system adopts the broad band lens assembly and the multi-spectral filter assembly to achieve the multi-spectral imaging system. The system is simple in structure and easy to achieve. Cofocal plane compensation in narrow band imaging is well achieved by thickness change of a band-pass parallel plate filter, and complexity of broad band lens design is reduced. The optical system is used for non-refrigeration broadband infrared multi-spectral imaging and has wide application prospect in the fields of military industry and civil use.

A multifunctional medical LED lighting system provided by the invention comprises four groups of LED modules which are exactly the same in sizes and provided with an R chip, a G chip, a B chip and a UW chip respectively; optical collimators with identical in structures are mounted at the front parts of the light-emitting surfaces of the LED chips respectively to collimate emergent light of the LEDs; three ways of emergent light of three LEDs of R/G/B are mixed into a one-way emergent light via an X-ray plate; two ways of emergent light of two LEDs of R/UW are mixed into a one-way emergent light via a dichroic mirror; the dichroic mirror is controlled by a motor to rotate for 45 degrees; when white light is chosen for lighting, the LEDs of G/B do not emit light, the LEDs of R/UW emit light, and at the moment, a UW (cold white) light and an R (red) light are mixed to generate a white light source via the dichroic mirror; when a RGB narrow-band light source is chosen, the UW LED does not emit light, and at the moment, the LEDs of R/G/B emit light at different time; when the UW LED goes wrong, the LEDs of R/G/B emit light at the same time and can be used as the spare white light source; and a condensing lens is mounted at the front of the dichroic mirror, so that the emergent light of the light source is enabled to be coupled in light fiber optic arrays conveniently. The multifunctional medical LED lighting system is compact in structure, low in cost, high in reliability, high color rendering white light and narrow-band images.

An observation device includes a light source unit capable of adjusting a quantity of 390-490 nm illuminating light according to a B1 range and a B2 range having a longer wavelength than the B1 range, and an imaging element that images reflected light from a subject irradiated with illuminating light. The light source unit changes the light quantity ratio of the B1 range to the B2 range depending on whether white light imaging is conducted or narrow band imaging is conducted. The imaging element includes RGB color filters arranged according to pixels and the R color filter transmits more light in an R range and the B1 range than light in the second B range and has a spectral characteristic that satisfies a predetermined conditional formulae.

The invention belongs to the technical field of medical image processing, and particularly relates to a deep detection network for quantifying esophageal mucosa IPCLs vascular morphological distribution. The deep detection network comprises a feature extraction network, a feature pyramid, a region candidate network, an interest region pooling and clustering distribution priori self-embedded cancerlesion classification network and a system for visualization on a narrow-band imaging endoscope image. The feature extraction network extracts a feature map of the input image; the feature pyramid fuses the features of different scales; the region candidate network proposes a possible lesion region; the region of interest is pooled, and the features are pooled to a suspicious lesion region; the cancer lesions are classified by a clustering distribution priori self-embedded cancer lesion classification network; and finally, visualizing is carried out on a narrow-band imaging endoscopic image,and frame selection marking is carried out on the cancer lesion by using different colors. The cancer focus of the early esophageal squamous cell carcinoma existing in the image is detected and diagnosed, the diagnosis efficiency can be effectively improved, and a doctor is assisted in obtaining higher diagnosis precision.

The invention provides an enhanced display narrow band imaging endoscopy system. The system comprises an image collection device, an image processing device and an image display device. The image collection device comprises multiple cameras, and the image processing device comprises an image collection module and a narrow band image processing module. According to the enhanced display narrow bandimaging endoscopy system, the cameras are used for collecting narrow band images and white light images, the white light images pass through the image collection module and then are directly output tothe image display device to be displayed in a normal mode, and the narrow band images are processed by the narrow band processing module and displayed in a narrow band mode through the image displaydevice. The narrow band mode display can improve the image quality of narrow band imaging and enhance the blood vessel image information. The invention further provides an imaging method of the enhanced display narrow band imaging endoscopy system.

The invention discloses a picture screening method and system for esophageal cancer model training, and a storage medium. The method comprises the following steps: inputting a to-be-screened static picture; clustering the static pictures by adopting a clustering algorithm according to the characteristics of the static pictures to obtain a plurality of types of static pictures; and screening the static pictures in each cluster by adopting a distance function to obtain static pictures with low similarity as training samples for establishing an esophageal cancer recognition model. According to the method, when the static pictures are input, a larger sample size can be allowed to be adopted to solve the problem of poor model generalization ability, meanwhile, the static pictures with large samples are clustered through a clustering algorithm, then the static pictures with low similarity in each cluster are screened through a distance function, and the static pictures with high similarity are obtained. And finally, on the premise that the sample coverage rate is not obviously influenced, the transformation from a large sample to a small sample is realized, and an esophageal squamous carcinoma lesion picture suitable for training and identifying narrow-band imaging is obtained.

Methods and apparatus for image processing suitable for use in wireless capsule endoscopy are provided. The image processing techniques exploit characteristic features of endoscopic images to enable low complexity compression. A color space conversion, coupled with lossless predictive coding and variable length coding are employed. Sub-sampling and clipping may also be used. The described image processing can be used both with both white-band imaging and narrow-band-imaging.

The invention belongs to the field of image recognition, particularly relates to a gastroscope image analysis system and method based on restoration and selective enhancement, and equipment, and aims to solve the problem that an existing gastroscope image recognition system cannot accurately recognize an early gastric cancer image. The method comprises the following steps: acquiring a narrow-band imaged gastroscope image as a to-be-detected image; performing preprocessing to obtain a to-be-detected image only containing gastric mucosa; obtaining a to-be-detected image without reflection through the reflection processing module; generating a composite image through the trained generative adversarial network, and automatically selecting a more vivid image; and enabling the trained gastroscope image recognition network to obtain the early gastric cancer probability of the image to be detected and obtain a region image of suspected early gastric cancer through a gradient weighted class activation mapping method. According to the method, selective enhancement of data features is carried out through the generative adversarial network, feature information most related to the classification task is automatically learned through the recognition model, and the accuracy of gastroscope image analysis is improved.

A method for sensitively diagnosing lower urinary tract disorders, particularly interstitial cystitis, is provided. The method can visually, simply, and clearly diagnose bladder ulcer without requirement of bladder hydrodistention under anesthesia. In the method, lower urinary tract disorders are diagnosed by observing an abnormality of a blood vessel and/or a newly formed blood vessel in a surface of bladder mucous membrane among abnormalities in blood vessels and/or newly formed blood vessels by visually comparing an image of the surface of the bladder mucous membrane and an image of a deep portion of the bladder mucous membrane obtained using a bladder endoscope system having a narrow band imaging device.

A semiconductor-based light source device for an endoscope that provides for a even power distribution and/or color distribution of light. The semiconductor-based light source device includes a plurality of semiconductor light sources and a plurality of fiber bundles, in which the respective fiber bundles are each segregated into a plurality of fiber sub-bundles for directing substantially the same amount of illumination from a respective semiconductor light source to each of a corresponding plurality of illumination windows disposed at a distal end of the endoscope. By providing such a semiconductor-based light source device, in which the respective semiconductor light sources are operative to adjust the intensity and/or color or wavelength of the light emitted by the respective semiconductor light sources, a desired color temperature and/or color balance for illumination of tissue and/or other matter under examination by the endoscope can be achieved.

The invention provides a tumor prediction method and device based on narrow-band imaging and an imaging endoscope, and the method comprises the steps: presetting a light source device and two image collection devices, and enabling the light source device and the two image collection devices to be arranged at a preset angle; comprising the following prediction steps that a light source device irradiates narrow-band spectrums of different colors to body tissues according to a preset irradiation frequency; the two image acquisition devices synchronously perform tissue image extraction on tissues in different colors according to different directions and the irradiation frequency to obtain a first tissue image and a second tissue image in different colors; identifying regions of interest of the first tissue image and the second tissue image to obtain a first region of interest and a second region of interest; obtaining prediction state information of the region-of-interest based on the first region-of-interest and the second region-of-interest; and calculating the prediction state information and outputting a prediction result.