The correct answer is Computer vision targets semantic understanding such as object or scene recognition, while image processing emphasizes pixel-level operations like filtering or edge detection. This option contrasts high level interpretation with low level pixel manipulations and it best captures the distinction the engineering team needs for landing zone detection and marker recognition.

Computer vision focuses on extracting meaning and structure from images and often relies on machine learning models to detect objects, classify scenes, or segment areas of interest. Image processing is used earlier in the pipeline to enhance images, reduce noise, or compute edges so that downstream computer vision models can work more reliably.

Image processing by itself lets the drone interpret delivery markers without any AI recognition models is incorrect because simple pixel operations can improve contrast or isolate shapes but they do not by themselves provide semantic interpretation. Recognizing symbols or text requires detection or classification models rather than only filters and transforms.

Computer vision only tweaks pixels and does not use machine learning is incorrect because modern computer vision commonly uses machine learning and deep learning to map pixels to labels, bounding boxes, or semantic masks. Pixel level tweaks are part of image processing and they support but do not replace CV models.

Computer vision and image processing are equivalent and interchangeable in drone scenarios is incorrect because the two fields operate at different levels of abstraction and they are complementary. Image processing handles low level enhancement and feature extraction and computer vision provides high level semantic understanding which is required for tasks like landing zone selection and hazard avoidance.

Full AWS Practitioner Certification Question