Fruit colour appearance: control the light before comparing the peel
Two real fruit comparisons show different recorded peel colours under the source natural-light and white-light conditions. They demonstrate why a colour trial needs controlled illumination and camera processing, rather than proving a grading result.
A colour threshold needs controlled capture and a defined reference.
Real camera capture
- Proposed light housing (spec)162 × 36 × 26 mm
- EvidenceReal application-lab captures
- Camera (proposed)2448 × 2048 colour
- Field (derived proposal)84.46 × 70.66 mm derived from printed 0.1x magnification
Lighting application-lab image sheet
What the camera saw
Cropped from the source test sheet; original test annotations are retained. Captures are not generated or enhanced. Drag a comparison slider, or open any image full size.
Source natural-light condition: Source natural-light lemon capture, with a different recorded peel appearance.
Source natural-light condition: Second fruit under the source natural-light condition; no grading result is implied.
The challenge
A peel-colour decision can change when lighting, white balance or exposure changes. Fruit curvature also produces highlights that may hide the useful colour region.
How it was solved
The source labels natural light and white light but does not name the exact lights, camera, exposure or white-balance settings. The proposed natural-light bars and colour camera are a separate controlled-trial setup.
The source calls one condition true colour. Without a calibrated reference and processing settings, these images support an appearance comparison rather than an objective colour-error result.
Why it works
- 1
A camera records the interaction of illumination, material reflectance and its own processing. These variables must be controlled before a colour threshold is trusted.
- 2
The supplied fruit photographs do not establish maturity, disease, variety, internal condition or a validated sorting rule.
Lighting application-lab image sheet
Evidence you can inspect
Both fruit examples change recorded colour between the two labelled lighting conditions.
The source provides appearance comparisons, not a measured colour difference, reference spectrum or calibrated produce-grading result.
An application-lab lighting demonstration, not a customer deployment or a qualified production inspection. Proposed camera, lens, fixture and scene spacing require a fresh trial. Source observations do not establish a defect-size limit, accuracy guarantee or production false-call rate.
From image contrast to a production decision
What to validate on your part
- 01
Set the reference
Use the actual produce, a suitable colour reference and fixed exposure/white-balance settings.
- 02
Choose the peel region
Exclude saturated highlights and test acceptable curvature, pose and surface variation.
- 03
Validate the grading decision
If colour supports sorting, compare against independently assigned labels across representative lots and maturity ranges.
From the demo to your test bench
Parts in this setup
LightOur pick
LS-4BRS15030-NL
Natural light, 162 x 36 x 26 mm (LS-4BRS15030-NL)
Two proposed broad-spectrum natural-light bars for adjustable illumination. Each printed housing is 162 × 36 × 26 mm; exact source light is not named. The 30 mm emitting width is printed; the 150 mm illustrated strip length is schematic and proposed, not a printed light-output specification.
$369.00
In proposed setup · quantity 2
CameraOur pick
MV-CS050-10GC
Hikrobot MV-CS050-10GC 5MP Area Scan Camera, GigE, IMX264, Color, Basic
Proposed 5 MP colour global-shutter camera, 2448 × 2048 pixels at 3.45 µm, C-mount. Colour processing, exposure and white balance must be controlled; the source camera is unidentified.
$739.00
In proposed setup
LensOur pick
LS-KF1628M-ML
16mm, F2.8, 1.1", C-mount, Macro lens (52 mm (-0.25X) to 546 mm (-0.03X)) (LS-KF1628M-ML)
16 mm C-mount macro lens. Its printed 0.1x / 152 mm working-distance pair gives a derived 84.46 × 70.66 mm field on the proposed camera. Confirm focus, coverage and fixture clearance.
$319.00
In proposed setup
Camera test stationOur pick
LS-BT-XZ-G-6-6045
Camera stand with XZ adjustment and glass tray, height 600 mm (LS-BT-XZ-G-6-6045)
600 mm column, 600 × 450 mm breadboard, glass tray and two lighting holders. Proposed bench fixture; actual load, travel, clamps, mounting fasteners and light clearance must be confirmed.
$359.00
In proposed setup
"Tested" identifies the part named in the source. "Our pick" is a proposed alternative or supporting part. Confirm colour, optics, controllers, power supplies, cables and mounting before ordering.
The set-up
| Source camera | Not identified in the source sheet |
|---|---|
| Camera and lens | Proposed catalog components |
| Proposed field | 84.46 × 70.66 mm derived from printed 0.1x magnification |
| Lens-to-part in proposed scene | 152 mm; printed working-distance pair |
| Mounting (proposed) | LS-BT-XZ-G-6-6045 · catalog test station with two lighting holders |
Mount the optics with intent
The station is an optical test bench, with its two lighting holders included in the catalog listing. The 3D column and base follow the published height and footprint. Carriages, clamps and part supports are schematic: confirm camera screw pattern, lens support, light fasteners, permissible load and actual travel before ordering.
Complete the electrical system
Select a controller against the exact light voltage, drive mode and power, then confirm camera data/power/I/O cables and a suitable PC. These are not silently added to the proposed optical kit.
Build this system
Start from this setup, adapt it to your part.
The configurator opens with this study's camera, lens, light and station picked. Swap any part, check fit and price, then buy or ask for a quote.
Camera and available lens CAD are published models. Light envelopes and the station base/height use printed dimensions. Part shapes, brackets, holders, shielding and spacing are illustrative; they are not reconstructions of a source part or a complete mechanically validated assembly. Light paths explain the proposed arrangement and do not simulate optical performance. The proposed field is derived from the printed camera pixel dimensions and the stated lens magnification/working-distance pair, not measured from the source capture. Specimen colours in 3D are illustrative display colours, not a calibrated colour target or a simulation of the light spectrum. The 30 mm emitting width is printed; the 150 mm illustrated strip length is schematic and proposed, not a printed light-output specification.
Food, pharma and agriculture
Related case studies
Food, pharma and agriculture
Contamination in blister packs
Reduce glare before searching for a dark speck.LightArch shadowless, 125 × 70 mm emittingExplore the study
Food, pharma and agriculture
Bread-piece outline and orientation
A usable outline is the first step; prove the handling task on real product variation.Light in sourceLS-2BS28238-RGBW · redExplore the study
Food, pharma and agriculture
Apple bruises: NIR vs SWIR
Match the wavelength, sensor and optics as one system.NIR940 nmExplore the study
Make the next trial specific.
Tell us the part size, smallest defect or tolerance, available working distance, material and line speed. Include what a good part and a reject look like.
About this case study: the photos and camera images come from the application laboratory of the maker of the lighting and optics we sell, imaged on the parts shown; your results depend on your part, camera and set-up. This is an application-lab study, not a LumaSpec customer deployment. Source figures are reported results; proposed components require a fresh trial. The 3D scene illustrates the arrangement and does not simulate imaging performance.