A pear sorting machine should be selected according to the colour differences and visible surface conditions that matter to the packhouse's customers. Camera specifications, AI features and machine speed alone cannot establish whether a system will make the required commercial grading decisions.
For pear packing operations, the key checks are variety-specific colour rules, inspection coverage around the stem and neck, acceptable natural skin markings, false rejection of marketable fruit and performance at the intended operating speed. Representative sample testing can help resolve these questions before the equipment configuration is finalised.
1. What Should the Pear Sorting System Be Asked to Classify?
Commercial pear grading involves more than separating attractive fruit from visibly damaged fruit. Packing programmes may apply different rules for colour, skin appearance, size and acceptable imperfections.
Define the inspection requirements around the grades the packhouse actually sells.
| Inspection item | Commercial purpose | What needs to be defined |
|---|---|---|
| Background colour | Separate fruit by accepted colour characteristics | Variety-specific colour references |
| Colour coverage | Evaluate visible surface colour | Coverage limits and tolerance |
| Natural russeting | Avoid rejecting normal skin patterns | Acceptable appearance by variety |
| Agreed surface conditions | Evaluate commercially relevant surface features | Visible feature, permitted extent and commercial consequence |
| Agreed visible handling-related marks | Identify relevant visible surface abnormalities | Acceptable and unacceptable marks |
| Shape | Assess conformity to grading requirements | Shape limits and presentation |
| Size or weight | Support commercial size categories | Required measurement modules |
A single label such as "defective pear" gives the inspection system too little information.
A brown surface pattern may be natural russeting on one variety but unacceptable under another commercial specification. The supplier needs examples of both conditions, labelled according to the buyer's grading rules.
The GELGOOG fruit sorting system provides a reference for intelligent grading technology. Pear-specific inspection functions require equipment evaluation and sample testing.
2. Why Are Pears Challenging for Optical Sorting?
Unlike uniformly round fruit, pears have a narrow neck, broad lower section and irregular shoulder profile. These characteristics affect fruit presentation and camera visibility.
Stem, Neck and Shoulder Coverage
A camera may capture the broad side of a pear clearly while part of the stem cavity or neck remains obscured.
Depending on the mechanical design, controlled fruit movement or multiple viewing positions may improve coverage. Camera quantity alone, however, does not establish whether commercially important areas are visible.
During evaluation, examine results for marks near the stem, neck and shoulder rather than relying only on defects located on easily visible surfaces.
Natural Russeting and Variety Differences
Pear skin appearance varies considerably between varieties.
Some pears have relatively smooth green or yellow skin. Others naturally develop brown russeting, visible lenticels or mixed surface patterns.
These features may resemble defects in captured images. Grading rules developed for one variety may not suit another.
A 2023 study of machine-vision pear grading examined shape, surface colour and defects in Korla fragrant pears using an experimental system. Its findings illustrate the importance of defined inspection methods, not GELGOOG equipment performance.
Visible Damage Is Not Always Detectable Damage
Certain handling-related marks may be visible during surface inspection. Hidden bruising, damage developing after inspection and internal-quality problems are different matters.
Firmness, internal browning and sugar content require separately specified measurement methods where applicable.
Define the visible conditions that matter commercially rather than assuming an optical camera can recognise every form of damage.
3. How Should a Packhouse Define Its Colour and Surface-Condition Rules?
Start with the packhouse's existing grading procedure. Quality staff can use current decisions about acceptable variation and customer-critical defects to prepare reference samples for the trial.
Prepare three groups of pears:
Clearly acceptable: Fruit meeting the intended grade, including normal variation.
Clearly unacceptable: Fruit with conditions excluded by the commercial specification.
Borderline: Fruit close to the acceptance boundary for colour or surface appearance.
Borderline examples are particularly useful. They expose disagreements between the machine's classifications and the packhouse's commercial expectations.
Build Variety-Specific Grading References
For each important pear variety, record its accepted colour range, natural skin characteristics, relevant surface conditions and customer-specific exceptions.
Photographs help communicate grading rules, although physical samples remain necessary for evaluating actual inspection performance.
Where two customers apply different standards to the same variety, record those differences separately.
The practical goal is consistent assignment to the packhouse's approved commercial categories, including fruit with natural appearance variation.
4. How Can You Tell Whether the Sorting Results Are Reliable?
A useful trial compares the sorter's decisions with labels agreed in advance by the packhouse and supplier.
Finding obvious defects is only part of the task. The packhouse also needs to know how frequently the machine rejects acceptable pears or assigns marketable fruit to an incorrect grade.
| Test result | Meaning | Commercial consequence |
|---|---|---|
| Correct acceptance | Fruit meets its assigned grade | Supports marketable output |
| Correct rejection | Fruit outside the agreed standard is separated | Protects packed-product quality |
| False rejection | Acceptable fruit is downgraded or rejected | May reduce commercial yield |
| Missed defect | Unacceptable fruit enters an approved grade | Creates quality risk |
| Wrong grade | Marketable fruit enters an incorrect category | Affects order consistency |
Why Overall Accuracy Is Not Enough
An overall accuracy figure can conceal weaker performance on commercially important surface conditions.
A system might classify common colour differences correctly but struggle with a particular skin condition that matters to premium-grade customers.
Ask for results by grade or surface-condition category, including false rejects and missed defects.
The report also needs to identify the tested varieties, reference labels, sample composition and operating conditions.
Protect Marketable Fruit from Unnecessary Rejection
False rejection deserves particular attention when incoming pears show substantial natural variation.
Include acceptable russeting, unusual but permitted colour patterns and borderline fruit in the evaluation.
Removing marketable pears from their approved grade reduces the commercial value recovered from a batch, even when defect detection appears effective.
An accuracy percentage becomes meaningful for procurement only when the test method, sample scope and acceptance thresholds are clear.
5. What Pear Samples Should Be Prepared for Testing?
A trial built from uniform, attractive pears will not answer the questions that matter during a mixed commercial run.
Include the following:
Normal production pears covering expected size, colour and skin variation.
Approved high-grade and secondary-grade fruit.
Known examples of commercially relevant visible defects.
Borderline pears close to acceptance limits.
Different varieties or storage conditions included in the operating scope.
Keep two sample purposes separate.
Representative batches show how the sorter behaves with a normal mixture of incoming fruit. Challenge samples concentrate on difficult surface conditions and borderline grading decisions.
A challenge set may contain a much higher proportion of defective fruit than normal production. Its results cannot establish routine commercial rejection rates.
Agree the sample quantity and reference classifications before testing. A small demonstration may show feasibility, but it does not show how the system will perform across normal production variation.
For multi-variety operations, include the differences most likely to affect classification rather than testing only the easiest variety.
6. Will Inspection Performance Hold at Production Speed?
Successful inspection at reduced demonstration speed does not establish performance during normal packing operations.
At the intended commercial throughput—for example, 2–3 tonnes per hour in a defined project—feeding quality, fruit spacing, inspection visibility and discharge timing can all affect results.
During the trial, watch three points closely.
Fruit Presentation
Pears need suitable separation and stable presentation when entering the inspection area.
Overlapping fruit, inconsistent spacing or uncontrolled orientation can interfere with classification. Test the actual range of pear sizes and shapes rather than a uniform selection.
Handling-Related Damage
Examine pears after feeding, conveying and discharge.
Transfer points, contact surfaces and outlet mechanisms may influence fruit condition. Use an agreed observation method and inspection timeframe when assessing additional handling marks.
Inspection and Routing Consistency
Evaluate grading decisions at the proposed operating rate, not just at demonstration speed.
Where multiple outlets are involved, check that graded fruit reaches its assigned destination without discharge congestion.
For outlet allocation, equipment integration and the distinction between nominal sorting capacity and packed output, see How to Plan 2–3 T/H Capacity and 12 Packing Outlets.
For existing packhouses, also check whether the current feeding and packing equipment can accommodate the proposed sorter.
7. What Should Be Agreed Before Equipment Acceptance?
Before final configuration or factory acceptance testing (FAT), the buyer and supplier need a shared description of the inspection task.
Document the fruit scope, grading references, operating conditions and evaluation method.
| Acceptance item | Required agreement |
|---|---|
| Fruit scope | Varieties, sizes and relevant sample conditions |
| Inspection scope | Colour rules and visible surface conditions |
| Reference grading | Approved classifications and borderline samples |
| Sample method | Representative batches and challenge sets |
| Performance measures | Grade agreement, false rejection and missed defects |
| Operating conditions | Target throughput and fruit presentation |
| Handling | Method for assessing additional mechanical damage |
| Reporting | Test results, limitations and unresolved conditions |
Acceptance thresholds should apply only to the agreed varieties, sample groups, operating speed and commercial grades. New varieties or defect categories may require additional testing.
Frequently Asked Questions
Can an optical pear sorting machine detect every bruise?
No. Some handling-related marks may be visible, but hidden or developing bruises require separate validation.
Can one grading programme work for different pear varieties?
Not necessarily. Colour, russeting and shape vary. Variety-specific settings may be required.
Is AI sorting always necessary?
No. Size-only or weight-based grading may not require advanced surface inspection. The commercial task determines the technology.
Can photographs replace physical samples?
No. Photographs help define requirements; physical samples reveal actual inspection and handling performance.
Get a Fruit Suitability & Sample-Test Checklist
If you are evaluating a pear sorting machine, share the varieties, grade descriptions, representative images, target surface conditions, incoming capacity and existing packing interfaces.
GELGOOG can use this information to prepare a fruit suitability and sample-test checklist before equipment configuration.
