100% pallet coverage with zero blind spots
Kompass AI Vision Platform with Integrated 2D Cameras + 3D Laser Inspection
Every Pallet. Every Defect. Inspected in 6 Seconds
Overview
A leading manufacturer of reusable wooden pallets supports global supply chains across manufacturing, warehousing, and logistics. Every pallet must satisfy strict structural quality standards to ensure safe handling, transportation reliability, and long service life.
Inspecting wooden pallets is considerably more complex than convention AI vision applications. Large pallet dimensions, multiple inspection angles, varying wood textures, and structural defects such as missing boards, broken components, exposed nails, twisted blocks, and damaged connectors make manual inspection both inconsistent and labour intensive.
To automate this process, the manufacturer partnered with Jidoka Technologies to develop an intelligent inline inspection platform that combines AI-powered 2D vision with 3D laser validation, delivering complete pallet inspection in a single automated workflow.
Opportunity
Maintaining consistent pallet quality at production speed introduced several operational challenges:
- Detect structural pallet defects in an inline process with a cycle time below six seconds.
- Achieve complete pallet coverage without blind spots.
- Inspect large wooden pallets from multiple viewpoints while the pallet remained on the conveyor.
- Reliably identify defects such as missing boards, free-standing nails, broken deck boards, twisted blocks, connector damage, missing components, and excessive overhang.
- Reduce operator dependency and eliminate variability associated with manual inspection.
- Develop a scalable inspection architecture suitable for deployment across multiple manufacturing sites.
The goal was to create a fully automated inspection solution capable ofdelivering consistent quality decisions while maintaining productionthroughput.
Jidoka's Approach
Hybrid 2D AI Vision + 3D Laser Inspection
Unlike conventional vision systems that rely solely on 2D imaging, Jidoka developed a hybrid inspection architecture combining 20 industrial 2D cameras with 3 large laser scanner . While the AI vision system identifies visible structural defects, the 3D laser data validates the physical geometry of the pallet, significantly improving inspection reliability.
2D data to identify location of free-standing nails. 3D data to measure position, height of nails. Zero Blind Spot Pallet Coverage
The multi-camera imaging arrangement provides complete visibility of the pallet's top, bottom, inner, and outer side surfaces. Combined with strategically positioned 3D lasers, the solution achieves 100% pallet coverage with virtually no blind spots, ensuring every critical structural component is inspected.
AI-Powered Structural Defect Detection
Images captured from multiple viewpoints are processed using Jidoka's deep-learning Kompass platform employing advanced object detection and segmentation models to identify defects including:
· Missing deck boards
· Broken boards across the width
· Missing wood sections
· Free-standing nails
· Twisted blocks
· Connector board damage
· Top deck holes
· Component overhang
· Missing structural components
· Insufficient nail joints
· Block damage and volume loss
The 3D inspection data further validates the detected defects, allowing operators to visualize defects identified in the 2D images directly on the corresponding 3D pallet model.
Intelligent Decision Making
Inspection thresholds can be tuned using a Human-in-the-Loop workflow, allowing production teams to optimize inspection sensitivity while minimizing false rejects. This adaptive approach delivers high throughput without compromising quality.
High-Speed Inline Integration
The complete inspection platform is installed directly over the manufacturer's existing conveyor line without requiring significant production modifications. The solution performs full pallet inspection in less than six seconds, enabling continuous inline operation while supporting future multi-site deployments through a standardized inspection architecture.