A TECHNICAL OVERVIEW OF LOGARITHMIC IMAGE SIGNAL PROCESSING ON ANDROID DEVICES

Asqar Ospan

2026

Full text Open access: Gold Metadata (CSL-JSON)

DOI 10.62731/mcnd-29.05.2026.010

Data confidence 70 medium

How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

This paper investigates the technical advantages of logarithmic (LOG) video recording on Android smartphones using the Camera2 API, focusing on dynamic range preservation and signal integrity compared to standard stock camera processing. An experimental comparison was conducted on a Samsung Galaxy S20 FE, capturing identical scenes in both Rec.709 and LOG formats. The analysis combines waveform evaluation with quantitative signal-to-noise ratio measurements, including both global and local SNR to assess noise behavior across luminance levels. The results show that while stock camera processing achieves higher global SNR due to aggressive noise reduction and tone mapping, LOG recording preserves a more uniform signal distribution and greater dynamic range, providing improved flexibility for post-production workflows.