Computer vision
Methods for extracting information from images and video, such as classifying objects, locating them or distinguishing image regions.
What does computer vision analyse?
Computer vision covers methods that extract information from images or video. Depending on the task, a whole image may be assigned to a class, an object's position determined, or an image divided into labelled regions. These tasks require different outputs and assessment criteria. An application for visible packaging damage therefore needs to define which damage it should identify, what images it receives and how its indications will be used in the workflow.
In a fictional dispatch process, a system is to examine package photos for possible corner damage and mark suspect areas. The team plans comparison cases with intact and damaged packages under different lighting conditions and camera angles. Both missed damage and false warnings are to be recorded. Visible damage to outer packaging alone does not establish damage to unseen contents. A responsible employee decides whether to hold or repackage the parcel based on the planned inspection. Unusable photos need a retake or another check within the process.
Sources
Related terms
- Machine learningModels learn patterns from data and apply them to new inputs.
- Deep learningMachine learning with neural networks containing multiple processing layers.
- Neural networkA model of connected computational nodes with adjustable parameters.
- AI inferenceApplying an already trained model to inputs to obtain a result.
- Multimodal AIAI using several information types, such as text, images or sound, in inputs or outputs.
- Optical character recognitionOptical character recognition makes writing in images machine-processable. Reading an amount is not the same as checking an invoice against business rules.
Articles on this topic
AI & Automation with ASES
AI assistants and automated workflows