Abstract: A method, and corresponding apparatus, for identifying a feature in an image comprises determining a plurality of characteristic values for each pixel or group of pixels within the image, and determining a confidence value for a target area of the image on the basis of the plurality of characteristic values of the pixels within the target area. Each of the plurality of characteristic values relates to a different characteristic of the feature. The confidence value is indicative of whether the feature is represented by the target area of the image.

Abstract: Disclosed is a non-transitory computer readable medium storing a computer program, in which when the computer program is executed by one or more processors of a computing device, the computer program performs operations to provide methods for detecting flaws, and the operations may include: extracting a flaw patch from a flaw image including a flaw; preprocessing at least one of the flaw image or non-flaw image not including a flaw; extracting a non-flaw patch from at least one of the preprocessed flaw image or non-flaw image; and training a neural network model for classifying patches to flaw or non-flaw with a training data set including the flaw patch and the non-flaw patch.

Abstract: Disclosed are systems, methods, and apparatus related to automated spectral selection for feature identification from remote sensed images. The invention includes various modules, such as a spectral selection processing module and a user device module that are communicatively coupled to each other via a communication connection. The invention includes a non-transitory memory that causes a processor to carry out one or more tasks. Those tasks include, but are not limited to, storing imagery; generating a list of imagery, input related to the imagery, a map view, target pixel values, and geometry related to the imagery; transforming pixels to vectors; conducting analytics within the identified vector features; quantifying data within identified vector features; and displaying the results on the user device module.

Abstract: A laparoscopic image smoke removal method based on a generative adversarial network, and belongs to the technical field of computer vision. The method includes: processing a laparoscopic image sample to be processed using a smoke mask segmentation network to acquire a smoke mask image; inputting the laparoscopic image sample to be processed and the smoke mask image into a smoke removal network, and extracting features of the laparoscopic image sample to be processed using a multi-level smoke feature extractor to acquire a light smoke feature vector and a heavy smoke feature vector; and acquiring, according to the light smoke feature vector, the heavy smoke feature vector and the smoke mask image, a smoke-free laparoscopic image by filtering out smoke information and maintaining a laparoscopic image by using a mask shielding effect. The method has the technical effects of robustness and ability of being embedded into a laparoscopic device for use.

Abstract: In some examples, a computer system may receive an image and associated location information corresponding to a location at which the image was captured. The system may provide the image as input to a machine-learning model previously trained to determine whether the received image includes a threshold amount of information to indicate that received location information indicates a location corresponding to the received image. Based at least in part on an output of the machine-learning model indicating that the received image meets the threshold, the system may store the associated location information as mapping information associated with the location.

Abstract: A system for establishing a contour of a structure is disclosed. An initialization subsystem (1) is used for initializing an adaptive mesh representing an approximate contour of the structure, the structure being represented at least partly by a first image, and the structure being represented at least partly by a second image. A deforming subsystem (2) is used for deforming the adaptive mesh, based on feature information of the first image and feature information of the second image. The deforming subsystem comprises a force-establishing subsystem (3) for establishing a force acting on at least part of the adaptive mesh, in dependence on the feature information of the first image and the feature information of the second image. A transform-establishing subsystem (4) is used for establishing a coordinate transform reflecting a registration mismatch between the first image, the second image, and the adaptive mesh.

Abstract: This application relates to a system for building machine learning or deep learning data sets for automatically recognizing geographical area information comprising a plurality of geographical area components provided on items. The system may include a first image database configured to store a first plurality of sets of images of geographical area information of items, each first set including an image of an entirety of geographical area information of an item. The system may also include a second image database configured to store a second plurality of sets of images of the geographical area information of the items, each second set including images of individual geographical area components. The system may further include a controller configured to convert the first plurality of sets of images into the second plurality of sets of images.

Abstract: Described herein are systems and methods of selecting treatment parameters values for treating skin lesions. A device may establish a treatment parameter selection model using a training dataset. The training dataset may include a plurality of examples. Each example may include a sample image of an example skin lesion to which a treatment is administered using an applicator. Each example may include a first label indicating success or failure of the treatment. Each example may include a second label corresponding to treatment parameters defining the treatment. The device may identify an input image. The device may determine that the input image corresponds to a skin lesion based on visual characteristics of the input image. The device may apply the treatment parameter selection model to the input image to output a recommended treatment to apply. The device may store an association between the input image and the recommended treatment.

Abstract: A sky filter method for panoramic images and a portable terminal is provided. The method comprises: obtaining several panoramic images containing sky regions as a data set, and marking a sky region and a ground region of each panoramic image to obtain a marked mask image and a panoramic template image of sky; extracting features of the sky region and the ground region of each panoramic image, and marking a positive sample and a negative sample; inputting the positive sample and the negative sample to SVM for training to obtain a model; extracting the features of the panoramic images, and inputting same to the model to obtain an initial mask image; removing a misclassified pixel and a misclassified region in the initial mask image to obtain an accurate mask image; and fuse the panoramic template image of sky and a test panoramic image to achieve a sky filter effect.

Abstract: Embodiments of the present disclosure generally relate to livestreaming methods and systems, and more particularly to whiteboard presentation systems that can be used in a livestreaming or video conferencing environment. In some embodiments, the whiteboard presentation system is configured to perform one or more processing operations, such as capture images on a whiteboard, perform image processing routines on the captured images, and transmit processed images as a video feed to one or more remote users, such as people or locations attending a video conference. The image processing routines can include one or more operations such as image denoising, contrast enhancement, color reconstruction, segmentation of a presenter, and image reconstruction.

Abstract: A method and system for characterization of Cannabaceae plants using macro photography images is disclosed. The method comprises the steps of receiving one or more macro photography images of a Cannabaceae plant; performing feature extraction analysis of trichomes using image processing, and performing plant characterization analysis using a neural network which analyzes the macro photography images. The training phase of the neural network comprises using results of chemical composition laboratory tests performed on the plants for which the macro photography images have been used in the training phase. The invention calculates and reports an assessment of maturity of the plant for harvesting, diagnosis of the existence of diseases, insects, or pests, assessment of the presence and concentrations of central ingredients, recommendations for treatment during plants drying, curing or storage production processes, and assessment of the quality and pricing of Cannabaceae plants products.

Abstract: Multiple images are continuously captured based on a photographing instruction. Multiple processed images without ghosts are obtained based on the multiple images. A fused image is obtained by inputting the multiple processed images without ghosts into an image fusion model. Noise of the fused image is lower than noise of any one of the multiple images, and a resolution of the fused image is higher than a resolution of any one of the multiple images. A target image is obtained and output by performing image enhancement processing on the fused image.

Abstract: Systems and methods are disclosed for performing automatic contour adaptation. The systems and methods receive first and second images depicting an anatomy of a subject and obtain a segmentation associated with the first image. The systems and methods apply a trained neural network to estimate the adapted segmentation corresponding to the anatomy depicted in the second image, the trained network consisting of two sub-networks: a registration sub-network, generating an initial segmentation estimate representing a deformation of the segmentation associated with the first image to fit the anatomy depicted in the second image, and a segmentation refinement sub-network, predicting a refined segmentation for the second image given the initial segmentation estimate.

Abstract: A method for deep learning video microscopy-based antimicrobial susceptibility testing of a bacterial strain in a sample by acquiring image sequences of individual bacterial cells of the bacterial strain in a subject sample before, during, and after exposure to each antibiotic at different concentrations. The image sequences are compressed into static images while preserving essential phenotypic features. Data representing the static images is input into a pre-trained deep learning (DL) model which generates output data; and antimicrobial susceptibility for the bacterial strain is determined from the output data.

Abstract: Provided is technology for extracting a person region from an image, that can suppress preparation costs of learning data. Included are a composited learning data generating unit that generates, from already-existing learning data that is a set of an image including a person region and a mask indicating the person region, and a background image to serve as a background of a composited image, composited learning data that is a set of a composited image and a compositing mask indicating a person region in the composited image, and a learning unit that learns model parameters using the composited learning data.

Abstract: Methods, non-transitory computer readable media, and arthroscopic video segmentation apparatuses and systems that facilitate improved, automatic segmentation analysis of videos of arthroscopic procedures are disclosed. With this technology, a video feed of an arthroscopic surgery can be automatically segmented using machine learning models and one or more tags related to the segments can be associated with the video feed. The generated videos can be output in real time to provide segmented information related to the surgical procedure or can be saved with the one or more segments tagged for playback for training or informational purposes.

Abstract: Certain examples provide an image data processing system including an anatomy detector to detect an anatomy in an image and to remove items not included in the anatomy from the image. The example system includes a bounding box generator to generate a bounding box around a region of interest in the anatomy. The example system includes a voxel-level segmenter to classify image data within the bounding box at the voxel level to identify an object in the region of interest. The example system includes an output imager to output an indication of the object identified in the region of interest segmented in the image.

Abstract: A skin detection method includes: dividing a region of interest in a face image into a highlighted region and a non-highlighted region; separately determining a first segmentation threshold of the highlighted region and a second segmentation threshold of the non-highlighted region; obtaining a binary image of the highlighted region based on the first segmentation threshold, and obtaining a binary image of the non-highlighted region based on the second segmentation threshold; fusing the binary image of the highlighted region and the binary image of the non-highlighted region; and identifying, based on a fused image, pores and/or blackheads included in the region of interest.

Abstract: An image analysis method is provided. In the image analysis method, a to-be analyzed image is inputted into a region-based convolutional neural network (RCNN) model to obtain a masked image outputted from the RCNN. The center of a masked object in the masked image is calculated. The center is regarded as an origin of coordinate and the farthest coordinate point from the origin of coordinate in each of the four quadrants relative to the origin of coordinate are searched. The image analysis block is generated for each of the farthest coordinate points. The post-processing is performed on the image analysis blocks to obtain an object range.

Abstract: A non-transitory computer-readable medium stores instructions readable and executable by a workstation (18) including at least one electronic processor (20) to perform an image reconstruction method (100). The method includes: determining a weighting parameter (13) of an edge-preserving regularization or penalty of a regularized image reconstruction of an image acquisition device (12) for an imaging data set obtained by the image acquisition device; determining an edge sensitivity parameter (?) of the edge-preserving algorithm for the imaging data set obtained by the image acquisition device; and reconstructing the imaging data set obtained by the image acquisition device to generate a reconstructed image by applying the regularized image reconstruction including the edge-preserving regularization or penalty with the determined weighting and edge sensitivity parameters to the imaging data set obtained by the image acquisition device.