A system and method relating to object detection may include receiving an image frame comprising an array of pixels captured by an image sensor associated with the processing device, identifying a near-field image segment and a far-field image segment in the image frame, applying a first neural network trained for near-field image segments to the near-field image segment for detecting the objects presented in the near-field image segment, and applying a second neural network trained for far-field image segments to the far-field image segment for detecting the objects presented in the near-field image segment.

Disclosed are an object positioning method, a video display method, an apparatus, a device, and a storage medium. The method includes: continuously intercepting a preset number of video frames in a video stream to be detected; detecting a first frame image in the intercepted video frames by a You Only Look Once (YOLO) object detection method to obtain a first positioning result corresponding to the first frame image; detecting each of other frame images in the intercepted video frames by a Kernel Correlation Filter (KCF) object tracking method according to a positioning result corresponding to a frame image prior to the each of the other frame images to obtain respective second positioning results corresponding to the other frame images; and keeping on continuously intercepting the preset number of video frames in the video stream to be detected and obtaining corresponding positioning results until the video stream to be detected is finished.

An image processing method and apparatus is provided for obtaining an image captured by an image capturing apparatus; determining characteristic data of the captured image based on inputting the obtained image into an image analysis model trained with images including a predetermined type of characteristic data; and identifying, based on the determination performed by the image analysis model, metadata including a numeric value regarding the predetermined type of the captured image.

Presented are concepts for feature identification in medical imaging of a subject. One such concept processes a medical image with a Bayesian deep learning network to determine a first image feature of interest and an associated uncertainty value, the first image feature being located in a first sub-region of the image. It also processes the medical image with a generative adversarial network to determine a second image feature of interest within the first sub-region of the image and an associated uncertainty value. Based on the first and second image features and their associated uncertainty values, the first sub-region of the image is classified.

A method and apparatus for detecting a liveness based on a phase difference are provided. The method includes generating a first phase image based on first visual information of a first phase, generating a second phase image based on second visual information of a second phase, generating a minimum map based on a disparity between the first phase image and the second phase image, and detecting a liveness based on the minimum map.

The present disclosure provide a method for identifying power equipment targets based on human-level concept learning, including: creating a dataset of power equipment images, and annotating power equipment in power equipment images; training neural network and Bayesian network with the annotated dataset and respectively acquire identification results and conditional probabilities; calculating probabilities of unions with the conditional probabilities; and filtering the identification result corresponding to the highest probability of the union as identification result of the dataset of the power equipment images and complete the identification of the power equipment. The present disclosure combines Mask R-CNN and probabilistic graphical model. The bottom layer uses Mask R-CNN, and the top layer uses Bayesian network to train in identifying power equipment images, so that a small amount of data samples can achieve good recognition, which improved the performance of Mask R-CNN model.

Systems and methods for image-based item identification are disclosed. Image data corresponding to one or more images depicting an item may be sent to one or more remote systems for image-based item identification. The identification indications and/or identification confidence scores received from the remote systems may be aggregated and weighted based at least in part on one or more factors related to the remote systems, the results, domains, image capture timing, image capture angles, and/or events to more accurately identify an item depicted in the images.

Provided are a training method for an air quality prediction model, a prediction method and apparatus, a device, a program, and a medium. The method includes the steps described below. A target monitoring range is divided into a plurality of regions; the air quality prediction model is pre-trained by adopting a pre-training sample and a pre-training objective function, where the pre-training sample includes measurement values; and the pre-trained air quality prediction model is trained by adopting a formal training sample and a formal training objective function, where the formal training sample includes the measurement values. The air quality prediction model is configured to predict air quality of the plurality of regions according to spatial information, historical information and environmental information.

The present disclosure relates to systems and methods for automatically identifying and verifying vehicle license plate data. Specifically, the inventive system utilizes multiple automated data points to correlate a more accurate read of a license plate taken on roadways than systems that rely solely on Optical Character Recognition. The inventive system utilizes machine learning to automatically determine the make and model of the vehicle, which is then matched against motor vehicle records to provide for automation in the 80-90% range. The inventive system also utilizes analytics to determine if there are issues with the cameras and provides for near real time alerts to maintenance personnel.

Computing systems may support image classification and image detection services, and these services may utilize object detection/image classification machine learning models. The described techniques provide for normalization of confidence scores corresponding to manipulated target images and for non-max suppression within the range of confidence scores for manipulated images. In one example, the techniques provide for generating different scales of a test image, and the system performs normalization of confidence scores corresponding to each scaled image and non-max suppression per scaled image These techniques may be used to provide more accurate image detection (e.g., object detection and/or image classification) and may be used with models that are not trained on modified image sets. The model may be trained on a standard (e.g. non-manipulated) image set but used with manipulated target images and the described techniques to provide accurate object detection.