GIS4035L is the laboratory for remote sensing and photo interpretation — working with aerial photographs and satellite imagery to extract information about the Earth's surface.
UWF's description covers both halves of the paired course: it "familiarize[s] students with the fundamentals of remote sensing and photo interpretation through hands-on techniques with aerial photographs and satellite imagery based on real-world applications," and is "broken up into two distinct sections — interpreting digital aerial photographs and examining and manipulating digital data from satellites and other remote sensors." ⚠ The laboratory specifically "will focus on techniques for the practical use of digital aerial photography and satellite imagery using both Erdas Imagine and ArcGIS Pro."
Florida's statewide description of the subject is broader and worth having: "principles of photographic and electromagnetic remote sensing systems which detect, record and measure distributions of natural and cultural phenomena. Interpretation of aerial and orbital imagery for urban and environmental research and planning."
The distinction between the two sections matters. ⚠ Photo interpretation is a visual skill — recognising what something is from its shape, size, tone, texture, pattern, shadow and context, with stereo imagery for relief. It is old, it is learned by doing it repeatedly, and it remains the ground truth against which automated methods are checked. Digital image processing is a numerical skill — imagery as a raster of measured radiance values in multiple spectral bands, manipulated mathematically.
⚠ The laboratory is where both are actually practised, and it is paired with the lecture GIS4035.
Two Florida public institutions carry this number: the University of West Florida and Florida State University, both at one credit.
| Institution | Its title | Credits | Paired lecture |
|---|---|---|---|
| University of West Florida (SUS) | Photo Interpretation and Remote Sensing Lab | 1 | ⚠ prerequisite GIS 4035 (3 cr) |
| Florida State University (SUS) | Introduction to Remote Sensing Lab | 1 | GIS 4035 (3 cr) |
Both carriers award one credit and pair it with a three-credit lecture — 4 credits and two grades for the subject. The 30 contact hours recorded here is derived: a one-credit laboratory conventionally carries two to three contact hours a week, so roughly 30 to 45 hours across a term.
⚠ The titles differ in emphasis and the difference is real, if small. UWF's Photo Interpretation and Remote Sensing signals that visual interpretation of aerial photography is a substantial component; FSU's Introduction to Remote Sensing is the broader, more common framing. ⚠⚠ Florida's statewide record titles this laboratory with its LECTURE's title — Remote Sensing of the Environment (U) — and gives it no separate description, so the state record cannot distinguish the two halves.
UWF places the course in the College of Science and Engineering, Department of Earth and Environmental Sciences.
⚠ Note the prerequisite structure differs between the two halves at UWF: the laboratory lists GIS 4035 (the lecture) as a prerequisite with a concurrency marker, while the lecture lists GIS 4043*/L*. Confirm with your adviser whether you take them together or in sequence, because the catalogue notation permits either reading.
Worth stating plainly because students naturally focus on producing a classification and treat assessment as a formality.
Any classification algorithm will produce a land-cover map. It will look authoritative, it will be colourful, and it will contain errors — systematically, in patterned ways. ⚠ Wet agricultural land is confused with wetland; shadow with water; bare construction with sand; young plantation with natural forest.
⚠⚠ What makes a classification a professional product is the accuracy assessment attached to it — an independent reference sample, an error matrix, and honest producer's and user's accuracy figures per class. A map without one is an unverified claim, and in Florida applications — wetland delineation for permitting, damage assessment for disaster funding, habitat mapping for regulation — the claim may be acted on legally or financially.
⚠ Learn to report what your product does not do well. A classification that is 92% accurate overall but poor at distinguishing two specific wetland classes is useful if you say so, and misleading if you do not.
Taken after the introductory GIS course and its laboratory — GIS4043/GIS4043L at UWF and FSU. ⚠ Note the number divergence for transfer students: the University of Florida numbers introductory GIS as GIS 3043C, a 3000-level integrated course, rather than a 4000-level lecture-plus-lab pair. Same subject, different number, different packaging — expect to show what you covered rather than matching on the number.
⚠ A 4000-level course carrying upper-division credit. Florida's statewide record classifies it as transferable to an institution offering the same course, with no Gordon Rule designation and no general-education category. It is marked for dual enrolment with elective high-school credit; ⚠ that marking is uniform across the entire GIS prefix, so it is boilerplate.
⚠ UWF dual-lists this laboratory with GIS 5027L, where graduate students are assigned additional work. Two consequences: the pace sits above a typical undergraduate lab, and taking the undergraduate version may block taking the graduate one for credit later. Ask before enrolling if you are considering a UWF master's.
⚠⚠⚠ Remote sensing is the part of geospatial science that machine learning has changed most, and by a wide margin. A guide to this course that treated AI as a side topic would be misdescribing the field a student is entering.
Image classification — the central technique of this laboratory — is a machine learning problem and always was. Supervised classification with training samples is supervised learning under an older name. What has changed is the methods: deep learning, and convolutional networks in particular, now substantially outperform traditional per-pixel classifiers on many tasks, because they use spatial context and texture rather than treating each pixel independently. ⚠ Esri ships deep learning tools inside ArcGIS Pro, and object detection from imagery is a production capability rather than a research one.
⚠⚠ The career implication is concrete and worth planning around. Manual digitising of features from imagery — tracing buildings, roads and land-cover polygons — was steady employment for a great many people, and it is being automated rapidly and effectively. Students should not aim at that work.
What is growing instead is exactly what this course's harder half teaches:
⚠ A specific caution about the underlying data, which practitioners take seriously. A model trained on imagery from one sensor, season, region or atmospheric condition frequently performs poorly on another — and the failure is silent, producing a plausible map that is wrong. In Florida this bites: seasonal water levels, dense vegetation, frequent cloud and rapid post-storm change all shift what imagery looks like. Validating on local, current data is not optional.
Practically, in this course: use these tools to have a concept re-explained, and to draft Python for processing workflows you then test. ⚠ Be wary of asking a language model for software procedures — Erdas and ArcGIS Pro are versioned and menu paths change, so generated instructions are frequently almost right and non-functional. Use the vendor documentation.
Generated September 15, 2026 · Updated September 15, 2026