Image integrity: Help us figure out the scale of the problem - Crossref

The Similarity Check Advisory Group met a number of times last year to discuss current and emerging originality issues with text-based content. During those meetings, the topic of image integrity was highlighted as an area of growing concern in scholarly communications, particularly in the life sciences.


This is a companion discussion topic for the original entry at https://0-www-crossref-org.library.alliant.edu/blog/image-integrity-help-us-figure-out-the-scale-of-the-problem/
1 Like

M3U8 Protocol uses Individual Video Files in a Folder and “Packaged,” (Compressed, Named, and Recognized…?)

Anyhow, if we took Pixel Units with nothing but Raw Pixel Data and ID within a Folder Compressed and processed by a “Viewer.” The Protocol could use a Registry Handler, to offer an open API that allows Runtime, Config, and Enviromental Registration of Pixel Specific Processing Functions/Algorithms/Etc… This way we offer, essentially, a Raw Format that is more-so flexible on the User-End device, with more Open-Well-Known status and less security risk. This also would eliminate the need for all the Conversions. It allows quite a few benefits that we currently forgoe for redundant BOM and Magic Packets…it used to be Similar…Way back.

Anyhow, the idea being simple; Try Catching a Function which Requires a Specific Format, defaulting with the Last Successful - and when Failing Cycling through the Set.

Then we could even Transcode individual Pixels entirely Separately - simple color? complex? images in images? Each within a system currently used to…simplify the security of copyrighted material, improve loading speeds and cacheability. It is a bit…on time. But if we skimp even most of the format and encoding packet information, in preference of a runtime - viewtime - processing, we could even locally cache outside the image - temporary folder? - the mechanisms used, which would even make curropted or integrity easier and more robust…the same image may be in a multiplicity of singatues…that is the issue?

Well, that is likely because of those issues; conversions, operating system retagging and features, this option keeps the system-unique separate and locale while maintaining the most customization per pixel on the market…why store 3 bytes for the color white or black when it only takes 1? Why not store a sequence of complex raw data that could be parameter fed into whichever function the most robust is registered by the viewing mechanism? Why, ABBA and Blurg (<- forgot the reference, predates me - but it is the original BMP…)

Love y’all <3
God Bless America. :innocent:

Implementing educational initiatives to raise awareness among researchers about the importance of maintaining image integrity and providing resources or training on how to handle visual data responsibly. Incorporating additional scrutiny into the peer review process to ensure the validity and accuracy of images presented in scholarly publications. This might involve the use of image analysis tools or the inclusion of experts in image interpretation. Exploring and adopting technology solutions that can aid in detecting image manipulation or misconduct. This may include the use of software tools designed to identify alterations or inconsistencies in visual data.
By addressing image integrity concerns, the scholarly community can uphold the standards of research and publication ethics, ensuring that visual data is accurately represented and supporting the overall integrity of scholarly communications.