Abstract: Provided are a method and electronic device for identifying a size of a measurement target object. The method includes imaging a reference object, which is a reference for identifying the size of the measurement target object, to acquire a reference object image, imaging the measurement target object to acquire a target object image, fusing the acquired reference object image and the acquired target object image, and inputting the fused reference object image and target object image to a first neural network model to acquire size information of the measurement target object from the first neural network model.

Abstract: A non-transitory computer-readable storage medium storing a determination program that causes at least one computer to execute a process, the process includes obtaining a plurality of pair images of a person obtained from a overlapping region of images captured by each of a plurality of cameras; generating a directed graph including nodes corresponding to person features obtained from each of a plurality of person images included in the plurality of obtained pair images; acquiring weights of links between the nodes in the generated directed graph based on a number of person images with similar person features between the nodes; and determining a combination of the person features in which a number of the person images with the similar person features in the plurality of pair images is maximized based on the acquired weights of the links.

Abstract: Systems and methods for tracking the location of a non-destructive inspection (NDI) scanner using scan data converted into images of a target object. Scan images are formed by aggregating successive scan strips acquired using one or two one-dimensional sensor arrays. An image processor computes a change in location of the NDI scanner relative to a previous location based on the respective positions of common features in partially overlapping scan images. The performance of the NDI scanner tracking system is enhanced by: (1) using depth and intensity filtering of the scan image data to differentiate features for improved landmark identification during real-time motion control; and (2) applying a loop-closure technique using scan image data to correct for drift in computed location. The enhancements are used to improve localization, which enables better motion control and coordinate accuracy for NDI scan data.

Abstract: Systems and methods are provided for classifying images associated with an item, and generating an image set for that item which includes image classifications determined to be helpful for the item type of the item. To classify images, an image classification model is generated and trained using two phases. The first phase uses intermediate model with text and visual processing to teach the model to recognize patterns created by text without requiring OCR at inference. The second phase uses visual processing to refine the model for use at inference. To generate an image set, image classifications helpful to an item type are identified, items are associated with item types, images are obtained for an item, the images are classified using the image classification model, missing image classifications set out in the preferred image set are identified, and a request or requests is generated for the missing image classifications.

Abstract: A method for detecting structures is provided. The method can include receiving inspection image data characterizing a region of interest of an object being inspected. The regions of interest can include one or more structures of the object. The method can also include determining, using a computer vision algorithm, a structure within the region of interest with respect to photometric properties of pixel data in the inspection image data. The structure can be determined using a predictive model trained to determine image filter parameter values for image filters of the computer vision algorithm based on applying optimization techniques using training image data and annotation data. An indication of the structure can be provided, for example for display or storage in memory. Systems and computer-readable mediums implementing the method are also provided.

Abstract: An object detection device has a controller configured to receive classification data and regions for an image. The regions are prioritized and if a region is selected for further prediction, a photographic feature for the selected region is determined and a re-capture of the selected region is made based on the photographic feature by updating an image capturing device. The re-captured region is re-processed and classification data is re-generated. Then, final object detection for selected region(s) is performed and final object detection for not-selected region(s) is performed.

Abstract: Sensing information indicating a state of a moving body is acquired. A plurality of captured images that capture a driver driving the moving body are acquired. From the plurality of captured images, captured images captured in a state of the moving body traveling forward based on the sensing information. A reference line of sight direction in a case of the driver facing forward is detected based on a statistical value of a line of sight direction of the driver indicated by each of the identified captured images.

Abstract: This application relates to a video encoding method, computer-readable storage medium, and computer device. The method includes: obtaining video data, the video data including a plurality of original video images; processing the video data according to a sequence generation rule to obtain a plurality of independent subsequence video image groups, each of the subsequence video image groups including subsequence images corresponding to the sequence generation rule; obtaining encoding manners corresponding to the subsequence video image groups; and independently encoding the corresponding subsequence video image groups according to the encoding manners to obtain subsequence video encoded data corresponding to each of the subsequence video image groups.

Abstract: The present invention relates to a guidance for treatment of a chronic total occlusion. In order to provide improved guidance information for chronic total occlusion treatment, a device (10) for guidance for treatment of a chronic total occlusion is provided that comprises an image supply (12), a data processor (14) and an output (16). The image supply provides a sequence of angiographic images comprising a vascular structure. The data processor detects at least one portion of the vascular structure indicating a total occlusion of a vessel based on the sequence of angiographic images; and determines an image of the sequence of images that shows at least one segment of the vessel next to the total occlusion; and generates guidance image data based on the determined image. The output provides the generated guidance image data. Thus, additional information relating to spatial aspects is provided to the user based on 2D image data.

Abstract: A method of inspecting a balcony identifies an area of the balcony where there is no access to pass a crawler robot from the outside of the balcony into the interior space of the balcony, drills a hole into the interior space of the balcony, passes the crawler robot that includes a humidity sensor through the hole into the interior space of the balcony. The method, by the processor of an electronic device, receives humidity measurements of the interior space of the balcony from the humidity sensor, compares the humidity measurements with a threshold value, determines that at least one of the humidity sensor measurements exceeds the humidity threshold value; and generates a notification indicating the humidity in the interior space of the balcony exceeds the humidity threshold value based on the determination. The method removes the crawler robot from the interior space of the balcony and seals the hole.

Abstract: There are provided a negative example availability deciding apparatus, a negative example availability deciding method, and a program that enable reuse of training data. A positive example evaluation training section generates a positive example evaluation classifier trained with target data as a positive example. A negative example evaluation training section generates a negative example evaluation classifier trained with target data as a negative example. An availability deciding section determines the classification accuracy of the positive example evaluation classifier and the classification accuracy of the negative example evaluation classifier by using evaluation data. The availability deciding section decides, on the basis of the classification accuracy of the positive example evaluation classifier and the classification accuracy of the negative example evaluation classifier, whether or not to use target data as negative example training data for training another classifier.

Abstract: Methods, systems, and apparatus, including medium-encoded computer program products, for pre-processing image data before 3D object tracking include, in at least one aspect, a method including: performing object detection in uncompressed, two-dimensional image data from a camera to produce two-dimensional location data for objects of interest; processing the two-dimensional location data for the objects of interest using a motion criterion to generate possible paths data for the objects of interest; and constructing a flight track of an object in three-dimensional space, from the possible paths data and position information obtained from a sensor, by filtering out false positives in the possible paths data.

Abstract: Various embodiments of apparatuses, methods, systems, and computer program products described herein are directed to a treatment system having a first group of in line emitters, a second group of in line emitters, and one or more payload sources, configured to identify a first group of one or more agricultural objects, select a first emitter from the first group of in line emitters to apply a first liquid-based treatment to the identified first group of one or more agricultural objects, identify a second group of one or more agricultural objects, select a second emitter from the second group of in line emitters to apply a second liquid-based treatment to the identified second group of one or more agricultural objects, simultaneously apply the first liquid-based treatment towards the first group of one or more agricultural objects and apply the second liquid-based treatment towards the second group of one or more agricultural objects.

Abstract: The present invention relates to a transmission image-based non-destructive inspecting method, a method of providing a non-destructive inspection function, and a device therefor. Specifically, when transmission images of industrial parts or electronic parts including an electronic circuit board mounted therein exist, the present invention enables various inspection methods to be used for the transmission images, so that a user can easily perform a defect inspection according to various situations.

Abstract: An apparatus and method for a machine learning engine for domain generalization which trains a vision transformer neural network using a training dataset including at least two domains for diagnosis of a medical condition. Image patches and class tokens are processed through a sequence of feature extraction transformer blocks to obtain a predicted class token. In parallel, intermediate class tokens are extracted as outputs of each of the feature extraction transformer blocks, where each transformer block is a sub-model. One sub-model is randomly sampled from the sub-models to obtain a sampled intermediate class token. The intermediate class token is used to make a sub-model prediction. The vision transformer neural network is optimized based on a difference between the predicted class token and the sub-model prediction. Inferencing is performed for a target medical image in a target domain that is different from the at least two domains.

Abstract: A method and system for treating a patient is described. The method includes: (i) receiving a sample from a patient, the sample including one or more cancer cells; (ii) obtaining, using an imaging device, one or more images of the cancer cells; (iii) processing, using an imaging processor, the one or more images to extract one or more image coefficients; (iv) mapping, using a trained classifier, the one or more image coefficients to a cancer cell type; (v) identifying, based on mapping the one or more image coefficients to a cancer cell type, one or more cancer cell types in the sample; (vi) identifying, based on the identified one or more cancer cell types in the sample, a course of treatment specific to the one or more cancer cell types; and (vii) treating the patient using the identified course of treatment.

Abstract: A computer implemented method for reading a test region of an assay, the method comprising: (i) providing digital image data of a first assay; (ii) inputting the digital image data into a trained convolutional neural network configured to output a first probability, based on the input digital image data, that a first region of pixels of the digital image data corresponds to a first test region of the assay; (iii) if the first probability is at or above a first predetermined threshold, accepting the first region of pixels as a first region of interest associated with the first test region; and (iv) estimating an intensity value of a portion of the first test region in the first region of interest.

Abstract: Provided are methods for automated verification of annotated sensor data, which can include receiving annotated image data associated with an image, wherein the annotated image data comprises an annotation associated with an object within the image, determining an error with the annotation based at least in part on a comparison of the annotation with annotation criteria data associated with criteria for at least one annotation, determining a priority level of the error, and routing the annotation to a destination based at least in part on the priority level of the error. Systems and computer program products are also provided.

Abstract: Methods and computing apparatus are provided for training AI/ML systems and use of such systems for performing image analysis so that the damaged parts of a physical structure can be identified accurately and efficiently. According to one embodiment, a method includes selecting an AI/ML system of a particular type; and training the AI/ML system using a dataset comprising one or more auto-labeled images. The auto-labeling was performed using the selected AI/ML system configured using a parts-identification model. The configuration of the trained AI/ML system is output as an improved parts-identification model.

Abstract: Approaches presented herein provide for semantic data matching, as may be useful for selecting data from a large unlabeled dataset to train a neural network. For an object detection use case, such a process can identify images within an unlabeled set even when an object of interest represents a relatively small portion of an image or there are many other objects in the image. A query image can be processed to extract image features or feature maps from only one or more regions of interest in that image, as may correspond to objects of interest. These features are compared with images in an unlabeled dataset, with similarity scores being calculated between the features of the region(s) of interest and individual images in the unlabeled set. One or more highest scored images can be selected as training images showing objects that are semantically similar to the object in the query image.