Computer Vision, Sensing, and Neuroscience

Computer vision

AI field that enables computers to interpret images/videos and make decisions from visual information.

Manufacturing defect detection

A computer vision use case where models inspect images of products for flaws.

Unstructured image data

Image/video data that does not fit a traditional relational table.

PyTorch

A deep learning framework commonly used to build neural networks, including computer vision models.

scikit-learn

A Python ML library often used for classical ML algorithms; less specialized for deep computer vision than PyTorch.

SQL

Language for structured relational databases; not the main tool for raw image modeling.

Pandas

Python tool for data analysis and tables; useful for EDA but not the core image-modeling framework.

Passive sensing

Gathering information without emitting a signal; example: camera capturing daylight.

Active sensing

Emitting a signal and measuring its return; examples: radar, ultrasound, echolocation.

Radar

Active sensing that emits radio waves and detects reflections.

Ultrasound

Active sensing using high-frequency sound waves.

Echolocation

Active sensing used by bats/dolphins; emitting sound and interpreting echoes.

Camera in daylight

Passive sensing because it records available light rather than sending out a signal.

Brain imaging

Observing brain activity directly; supports cognitive science and AI-related understanding of cognition.

Cognitive bias

Systematic pattern in human thinking/decision-making; a psychology concept, not the same as direct brain imaging.