--- title: "3D Observation Glossary" slug: "3dobservation-glossary" tags: ["Observation Scan", "3D Localization", "AIR3D", "3D", "3D Observation"] updated: 2026-10-27T23:12:37Z published: 2026-10-27T23:12:37Z canonical: "ims-handbook.cenosco.com/3dobservation-glossary" --- > ## Documentation Index > Fetch the complete documentation index at: https://ims-handbook.cenosco.com/llms.txt > Use this file to discover all available pages before exploring further. # 3D Observation Glossary

Below you will find the definition / meaning for the relevant AIR3D terminologies. Take note that all AIR3D related acronyms are included in the IMS Abbreviations list.

 ## A | Term | Meaning | | --- | --- | | AIR | AIR (Autonomous Integrity Recognition) is the algorithm that is being used by Machine Vision to recognize the integrity issues on the Images. | | AIR Data Engineers | This is the AIR3D Users that upload 2D Images & Point Clouds into IMS, run Machine Vision to detect integrity issues, do initial review of the Images, run 3D Localization to map OB-Items to 3D models & Equipment, and create OB-Projects. See [Observation Scan](/v414/docs/observation-scan-1) and [3D Localization](/v414/docs/3d-localization). | | AIR Inspector | This is the AIR3D Users that review the OB-Items in IMS, verify in Field if needed, and create ECHs & CAIRs. See [3D Observation](/v414/docs/3d-observation). | | Annotations | These are rectangles drawn by Machine Vision or the AIR3D Data Engineer on 2D Images in the Observation Scan Module, where an integrity issue is identified. Annotations become Observation Items (OB-Items) through the 3D Localization process. | ## B | Term | Meaning | | --- | --- | | Batch | In the Observation Scan List, any "last level" folder can be [split into Batches](/v414/docs/machine-vision#splitting-into-batches). This basically means that a set of Images (group of photos) is split into smaller sets of Images and these subset are the Batchs. | | Box | In the 3D model view, Observation Items are displayed as boxes. The boxes can be resized and moved when reviewing Observation Items. | ## C | Term | Meaning | | --- | --- | | CAIR | A CAIR is a Corrective Action Schedule. If a Corrective Action is needed, this can be indicated on the OB-Item. Certain Integrity API Classes will require a CAIR. | | Completed Status | Observation Items that have been added to an ECH get the Completed Status. An Observation Project is Completed when all OB-Items in the Project has either the Completed or Obsolete Status. | ## E | Term | Meaning | | --- | --- | | ECH | Equipment Condition History. All OB-Items should be added to an ECH. | ## F | Term | Meaning | | --- | --- | | FILTERING Stage | One of the [Machine Vision Stages](/v414/docs/machine-vision#setting-the-machine-vision-stages). In this stage similar Images are filtered. | | Filter Images Setting | The Filter Images [Observation Scan Setting](/v414/docs/machine-vision#setting-the-machine-vision-stages) will determine if the results of previous Machine Vision runs will be reused. | | FILTERING PATCHES Stage | One of the [Machine Vision Stages](/v414/docs/machine-vision#setting-the-machine-vision-stages). This stage filters out similar Image patches. This is only used if the Images are split up. See Segmentation Factor. | ## I | Term | Meaning | | --- | --- | | Inherit | This is referring to the inheritance of Settings. Observation Scan Settings can be specified for each Tag in the Observation Scan List. If not specified, they will inherit it from the level above. | | Image Flattening | This [converts panoramic Images into planar (flat) Images](/v414/docs/reviewing-images#converting-360-panorama-images-into-planar-images). Machine Vision works best on planar Images. Image Flattening should not be applied to planar images as it will distort them. | | Image Gallery | You can view the Images (before and after the Machine Vision run) in the Image Gallery. Here you can delete unwanted Images and review the Machine Vision Annotations. [The Image Gallery can be activated from the Observation Scan Main Screen](/v414/docs/review-images#opening-images-in-the-image-gallery). The AIR Inspector can also open an OB-Item's Images in the Image Gallery from the [OB-Item List](/v414/docs/3d-observation-item-list) by selecting an OB-Item and clicking the **Image Gallery** button. | | Integrity Issue | Issues that can lead to an increased risk of failure in Equipment. This is what Machine Vision identifies as Annotations on the Images. | | Issue API Class | The Issue Class describe the Integrity Issues on Equipment, such as corrosion, damaged insulation, etc. In the Observation Scan, Machine Vision identifies integrity issues on the Images. The integrity issues are classified based on the Issue Classes and marked as Annotations. An OB-Item is created from an Annotation and the Issues Class, now called Issue API Class, forms part of the OB-Item's properties. | ## L | Term | Meaning | | --- | --- | | Localization | 3D Localization is the process of calculating the 3D positions for 2D Annotations. | ## M | Term | Meaning | | --- | --- | | Machine Vision | Machine Vision makes use of the AIR algorithm to detect integrity issues for the provided Images. These issues will be marked as Annotations on the Image and will be classified with an Issue Class. These Annotations become the OB-Items in the Observation Project. | | Merging Projects | You can [merge](/v414/docs/3d-observation-list#merging-projects) multiple Observation Projects. The Observation Items in the Observation Project will then be combined in one Observation Project. | ## O | Term | Meaning | | --- | --- | | Observation Scan Settings | Observation Scan Settings can be specified for each Tag in the Observation Scan List. These settings will determine which [Issue Classes](/v414/docs/machine-vision#selecting-which-issue-classes-to-report) are detected, as well as the [Machine Vision Stages](/v414/docs/machine-vision#setting-the-machine-vision-stages). It also determines [Image Flattening](/v414/docs/reviewing-images#observation-scan-settings-for-image-flattening), i.e., how 360° panorama Images are converted into planar Images (according to the FOV (Field of View), Pitch, and Yaw). | | OB-Item | Same as Observation Item. See below. | | Observation Item | Machine Vision makes Annotations to Images if it identifies Integrity Issues. These Annotations are then mapped to the 3D model during 3D Localization. It then becomes OB-Items that include a mapped Equipment and Issues API Class. The OB-Items are shown as boxes on 3D model. | | OB-Project | Observation Project. See below. | | Observation Project | An Observation Project is a list of Observations Items, mapped to the 3D Model created through the Observation Scan process. | | Observation Project Status | The Observation Project can have different Statuses (Unresolved, Created, In Progress, Completed). When the AIR Data Engineer Resolves the Project the Status will change to "Created". Thereafter the Status is automatically updated based on the Status assigned to each OB-Item within the Project. The Status will go to "Completed" when all OB-Items in the Project has been marked as either "Completed" or "Obsolete". | | Observation Scan | Observation Scan is an IMS Module that allows you to import Images into IMS, use Machine Vision on the images to detect integrity issues, and review and prepare the images for 3D Localization. | ## P | Term | Meaning | | --- | --- | | Panoramic Images | Images that are often captured by a special lens or by rotating a lens to capture a wide field of view. Such images are distorted toward their edges. Panoramic Images should be flattened for use with Machine Vision. | | Photo Group | A subgroup of photos used in the Observation Scan module. Photo groups can be broken up into smaller groups referred to as "Batches" through the Split function in the Observation Scan List. | | Planar Images | Images with a relatively narrow field of view such that the distortion between the center of the image and the edges is small. | | Point Cloud | A Point Cloud is a discrete set of data points in space. The points may represent a 3D shape or object. Each point position has its set of Cartesian coordinates (X, Y, Z). A 3D scan is performed to capture the Point Cloud. Point Clouds are imported during Observation Scan process by the AIR Data Engineer. | ## R | Term | Meaning | | --- | --- | | Resolving a Project | When all 3D positions are calculated, and all images are reviewed, the AIR Data Engineer should [Resolve](/v414/docs/resolve-3d-project) the Observation Project. The Project Status will then change to "Created". This means the it is ready for the the AIR Inspector can start reviewing the OB-Items. | | Reviewing Inspection Findings | This is done by the AIR Inspector. He needs to review all Observation Items (OB-Items) in his Observation Project, to ensure that they are linked to the correct Equipment and have the correct Issue API Class selected. | | Reviewed Images (during Observation Scan) | In [Observation Scan](/v414/docs/review-images), the Machine Vision results should be reviewed by the AIR Data Engineer. When he is satisfied he should mark the Images as "Reviewed". When an Image has been marked as Reviewed, all non-modified Machine Vision Annotations will be flagged as correct. | | Reviewed Images (during 3D Localization | During 3D Localization the AIR Data Engineer can [set the Review Status](/v414/docs/review-3d-calculations#setting-images-as-reviewed) on the Image details to Reviewed (blue), after he reviewed the 3D Localization calculation results, i.e., the alignment of the Annotations. When all Images have been reviewed, he should Resolve the Project. This means his work is done and the AIR Inspector can start reviewing the OB-Items. | | Reviewed OB-Item | The AIR Inspector should review all OB-Items in the OB Project, i.e., check if correct Equipment and correct Issue API Class are selected. He can then set the OB-Item Status to Reviewed. Only Reviewed OB-Items can be added to an ECH. | ## S | Term | Meaning | | --- | --- | | Segmentation Factor | The Segmentation Factor is a [Observation Scan setting](/v414/docs/machine-vision#setting-the-machine-vision-stages). It should be used when high resolution Images are provided. Machine Vision tends to return better results if high resolution Images are split into smaller Images. | | Stages | Machine Vision runs in one to three Stages (FILTERING, FILTERING PATCHES, MACHINE VISION), depending on the Filter Images setting and the Segmentation factor value in the [Observation Scan settings](/v414/docs/machine-vision#setting-the-machine-vision-stages). | ## T | Term | Meaning | | --- | --- | | Tag | Each row in the [Observation Scan](/v414/docs/importing-images) List has a Tag name. It could be a folder, subfolder, group of photographs (Images), or a Batch. | ## U | Term | Meaning | | --- | --- | | Unresolved Observation Project | When the 3D Localization work is still in progress, the Project Status will be "Unresolved". |