Systems and methods for training an artificial intelligence (AI) classifier of scanned items. The items may include a training set of sample raw scans. The set may include in-class objects and not-in-class raw scans. An AI classifier may be configured to sample raw scans in the training set, measure errors in the results, update classifier parameters based on the errors, and detect completion of training.

Systems and methods are disclosed for selecting target objects within digital images utilizing a multi-modal object selection neural network trained to accommodate multiple input modalities. In particular, in one or more embodiments, the disclosed systems and methods generate a trained neural network based on training digital images and training indicators corresponding to various input modalities. Moreover, one or more embodiments of the disclosed systems and methods utilize a trained neural network and iterative user inputs corresponding to different input modalities to select target objects in digital images. Specifically, the disclosed systems and methods can transform user inputs into distance maps that can be utilized in conjunction with color channels and a trained neural network to identify pixels that reflect the target object.

A computing system includes a processor; and a memory having stored thereon an adjustment application comprising computer-executable instructions that, when executed, cause the computing system to: display a graphical user interface including a digital medical image of a patient; superimpose a bounding box; receive an adjustment of an area of interest; and provide an adjusted digital medical image. A non-transitory computer-readable medium includes computer-executable instructions that, when executed via one or more processors, cause a computer to: display a graphical user interface including a digital medical image of a patient; superimpose a bounding box; receive an adjustment of an area of interest; and provide an adjusted digital medical image. A computer-implemented method includes: displaying a graphical user interface including a digital medical image of a patient; superimposing a bounding box; receiving an adjustment of an area of interest; and providing an adjusted digital medical image.

An image processing system includes a detection processing portion, a first correction processing portion, and a reading processing portion. The detection processing portion detects an outline of an information code included in a captured image. The first correction processing portion corrects a shape of the information code based on the outline of the information code detected by the detection processing portion. The reading processing portion reads information included in the information code from the information code after correction by the first correction processing portion.

A data marking method includes obtaining a to-be-recognized image that includes to-be-recorded data, the to-be-recorded data being displayed by a display apparatus, recognizing the to-be-recorded data in the to-be-recognized image, detecting the data content displayed by the display apparatus according to the to-be-recorded data, and displaying a marking according to a detection result.

A computer-implemented method includes preprocessing a variable dimension medical image, identifying one or more areas of interest in the medical image; and analyzing the one or more areas of interest using a deep learning model. A computing system includes one or more processors; and one or more memories storing instructions that, when executed by the one or more processors, cause the computing system to preprocess a variable dimension medical image, identify one or more areas of interest in the medical image; and analyze the one or more areas of interest using a deep learning model. A non-transitory computer readable medium contains program instructions that when executed, cause a computer to preprocess a variable dimension medical image, identify one or more areas of interest in the medical image, and analyze the one or more areas of interest using a deep learning model.

A method for image capture using artificial intelligence (AI) techniques to generate a user-personalized and noise-corrected final image from a captured image, that includes classifying a noise associated with a lens of an imaging device, preprocessing the captured image based on the classified noise to determine an initial region of interest (ROI), generating a first processed image by inputting the initial ROI and the captured image to a deep learning network, receiving a passive user input corresponding to a portion of a first preview of the first processed image, determining an additional ROI based on the passive user input and the classified noise, generating a second processed image by inputting the second ROI and the captured image to the deep learning network, and generating a user-personalization based noise-corrected final image based on the second processed image.

Systems and methods are disclosed for selecting target objects within digital images utilizing a multi-modal object selection neural network trained to accommodate multiple input modalities. In particular, in one or more embodiments, the disclosed systems and methods generate a trained neural network based on training digital images and training indicators corresponding to various input modalities. Moreover, one or more embodiments of the disclosed systems and methods utilize a trained neural network and iterative user inputs corresponding to different input modalities to select target objects in digital images. Specifically, the disclosed systems and methods can transform user inputs into distance maps that can be utilized in conjunction with color channels and a trained neural network to identify pixels that reflect the target object.

The present disclosure relates to systems, non-transitory computer-readable media, and methods for training and utilizing scale-diverse segmentation neural networks to analyze digital images at different scales and identify different target objects portrayed in the digital images. For example, in one or more embodiments, the disclosed systems analyze a digital image and corresponding user indicators (e.g., foreground indicators, background indicators, edge indicators, boundary region indicators, and/or voice indicators) at different scales utilizing a scale-diverse segmentation neural network. In particular, the disclosed systems can utilize the scale-diverse segmentation neural network to generate a plurality of semantically meaningful object segmentation outputs. Furthermore, the disclosed systems can provide the plurality of object segmentation outputs for display and selection to improve the efficiency and accuracy of identifying target objects and modifying the digital image.

Examples are described of digital zoom retaining image characteristics such as sharpness, clarity, and/or contrast. In some aspects, a device can receive an image and can determine various image characteristic scores corresponding to digitally zoomed variants of the image having different zoom strengths. For instance, the device can determine a first image characteristic score for a first zoom strength and a second image characteristic score for a second zoom strength. The device can compare the image characteristic scores to an image characteristic threshold, and can select the highest zoom strength for which the corresponding image characteristic score is not below the image characteristic threshold. For example, the device can select the first zoom strength if the first image characteristic score meets or exceeds the image characteristic threshold while the second image characteristic score does not. The device can output image data corresponding to the image at the selected zoom strength.