Glossary¶

This glossary defines key terms used throughout the Robovision AI documentation.

C¶

Classes: Classes are categories of objects that you need to differentiate in the samples. It can be any quality like ok or nok or a noun like cat or dog. In classifications, an entire image is one class; in other algorithms, classes are assigned to multiple items present in an image.
COCO: COCO imports consist of images and their associated annotations saved in a specific JSON structure. This format defines how classifications, bounding boxes, masks, and metadata (such as image height and width) are stored.

Data points: Data points are individual measurements plotted on the line chart, each representing a specific moment in time with its associated metrics.
Datasets: Datasets are sets of images and annotations used for training or testing a model. Datasets contain one annotation per sample. Datasets cannot be edited after they are created, protecting the data within, so you can repeat a training or a test and get the same results. Datasets are created in the Datasets panel of the label center. If you train a model using "auto setup", two datasets will automatically be created in the process.

Empty annotations: Empty annotations are used when a sample doesn't contain any of the classes present in the project. Assigning an empty annotation indicates that none of the existing classes apply to the sample.

Image directory-per-class: Image directory-per-class is a data format where the imported samples must be organized into one folder per class, so that each folder contains images of an object or scene belonging to the same class. According to the name of the folder, the annotations of a label type are added to each of the images.
Imports: Imports are data folders imported into a project and displayed in the Label center section of the project details page. In the Filter panel of the label center, imports can be used to filter samples.

KRF or KLARF (KLA Tencor Result File): KRF or KLARF (KLA Tencor Result File) is a file format used to transfer data about defects in the semiconductor industry. KRF is a text file that can contain information about the wafer, references to external image files, data about the defects location, and the list of classes. Robovision AI supports the 1.2 and 1.8 versions of KRF files.

Label: Label means to assign classes and make selections in samples. The terms label and annotate are used interchangeably.

Major: Major versions typically introduce significant changes that might not be compatible with previous versions. In version numbering, a major version is indicated by the first number.
Minor versions usually involve smaller updates that: Minor versions usually involve smaller updates that are backward-compatible with the major version. For example, if an algorithm is updated from version 5.7 to 5.8, it generally means that improvements have been made without changing how the algorithm fundamentally works. The minor version is represented by the second number in version numbering.

OPC-UA (Open Platform Communications Unified Architecture): OPC-UA (Open Platform Communications Unified Architecture) is a secure, standardized protocol for industrial communication. It enables seamless data exchange between devices and software applications in automation systems. OPC-UA is used to provide real-time inference results from Robovision AI to external systems.

Pascal VOC: Pascal VOC is an XML file, unlike COCO which has a JSON file. In Pascal VOC, we create an XML file for each image in the dataset, while in COCO we have one file each, for entire dataset for training, testing, and validation.

Sources: Sources refer to who or what made an annotation. Annotations have different origins—some are labeled by members, while others are predicted by algorithms with models. In other words, a source is where the annotation came from.

Tags: Tags are useful if you need to assign custom characteristics to samples. Tags act like filters—you select a tag, and then it filters the samples by that tag. You can combine selecting filters and tags.
Test: Test is a process of evaluating a model.

Webhook: Webhook is a mechanism for transmitting real-time data from one application to another in response to specific events. This way, every time a camera inference makes a prediction, Robovision AI automatically communicates with a third-party service or custom application via HTTP POST requests.