Capture-Based Scientific Imaging: What Makes an Image Scientific?

By Helvy Giovanny Sierra Vargas, scientific photographer and lecturer in scientific imaging, Universidad del Rosario (Colombia)

 

Instagram: @helvysierravargas

An Introduction to Capture-Based Scientific Imaging (CBSI)

At the time of writing this article (April 2026), a crewed mission once again places humanity in lunar orbit. Beyond the technological milestone, one element has regained central importance: images. Not as aesthetic pieces or mere records, but as means of accessing phenomena that, in many cases, had not been observed under these conditions for decades. This context brings back a question that is often overlooked: What makes an image scientific?

From image as representation to image as data

When we think about image recording, we usually approach it from an aesthetic or communicative perspective: we seek impact, composition, and visual narrative. Even in documentary contexts, the image serves the function of reporting an observable event.

However, in science, the image ceases to be merely representation and begins to operate as data. This shift is not minor. It implies that the image is no longer evaluated only by what it shows, but by how it was produced, under what conditions, and with what level of reliability it can be interpreted.

Three levels of image construction: aesthetic, documentary, and scientific

To illustrate this difference, we can think of three levels of image construction:

· Aesthetic/creative image: prioritizes composition, visual impact, and the author’s interpretation.

· Documentary image: aims to faithfully record an observable phenomenon, minimizing creative intervention.

· Scientific image: not only records, but is produced under conditions that allow validation, analysis, and integration into a knowledge production process.

This is not a hierarchy of value, but of function. A documentary image can be rigorous, but it does not necessarily meet the criteria that allow it to operate as scientific data. To understand this difference, it is not enough to describe it; it is necessary to observe how the same scene can be constructed under different criteria.

Figure 1. Aesthetic construction of the image     

Figure 2. Documentary record of the phenomenon

Figure 3. Scientific image under controlled capture conditions

The difference between these three images does not lie only in their appearance, but in the conditions under which they were produced. While the first prioritizes aesthetic decisions and the second seeks descriptive fidelity, the third incorporates technical information, control of variables, and capture conditions that allow analysis and validation.

So, does a scientific image emerge or is it constructed? The answer is: both. An image can emerge from a circumstantial event (for example, an unforeseen phenomenon) and later acquire scientific value if it can be subjected to validation processes. It can also be the result of rigorous planning, where capture protocols, variable control, and recording conditions are defined from the outset. What matters is not the origin, but the possibility for the image to be sustained as evidence.

Capture-Based Scientific Imaging (CBSI)

In this context, Capture-Based Scientific Imaging (CBSI) can be understood as a type of visual record whose validity does not depend on its aesthetic value or solely on its documentary nature, but on its capacity to be integrated as data within a knowledge production process through conditions of traceability, verifiability, and analysis. This definition introduces a key shift: the image is not the final result, but a unit within a system of knowledge.

Criteria that make an image scientific

For an image to operate as CBSI, it must meet at least three fundamental conditions.

· Traceability: It must be possible to identify the origin of the image, including capture conditions, equipment used, technical parameters, and production context. Without this information, the image loses its ability to be validated.

· Verifiability: The image can be analyzed by others and contrasted with the observed phenomenon. It does not depend solely on the author’s interpretation.

· Reproducibility (when applicable): In some cases, the phenomenon can be replicated under similar conditions. In others (such as unique events) what is reproducible is not the phenomenon itself, but the conditions of the recording and its methodological consistency. This distinction is essential: not all scientific images come from repeatable phenomena, but they must come from controllable and transparent processes.

Beyond recording: intention and method

A common mistake is to assume that scientific imaging simply consists of recording an event. It does not. CBSI implies intentionality and method. Even when the event is emergent, the way the image is interpreted, analyzed, and contextualized determines its scientific value. This requires a shift in perspective: moving from capturing what happens to constructing visual evidence under scientific criteria.

Technology, development, and validation

This same principle has enabled the development of the tools we use today. Cameras, tripods, filters, and support systems are not the result of arbitrary decisions, but of iterative processes based on hypothesis, testing, and validation.

In this sense, technological development in photography (including brands such as K&F Concept) can also be understood as a process analogous to the scientific method: questions are formulated, solutions are designed, tested, discarded, or refined. The tool, therefore, is not neutral; it is the accumulated result of validated knowledge.

Closing remarks

Thinking about images from a scientific perspective implies shifting the question from how it looks to how reliable it is as evidence. Capture-Based Scientific Imaging is not defined by its appearance, but by its ability to be sustained within a knowledge production process.

In the next article, we will explore one of the most critical variables in this process: light, not as an aesthetic resource, but as a structural condition of visual data quality.