These cases are organized by application scenario, technical modules, deliverables and delivery boundaries, helping customers assess similar project routes, required materials and acceptance criteria.
For RK3588 Android smart glasses, Wenge delivered offline frontal/profile face detection, Android AAR SDKs, a Camera2 demo project, integration notes and checksum scripts. The case does not claim identity recognition, face libraries or person search capability.
The project addressed temperature acquisition, camera storage, preview latency and write-confirmation issues in a semiconductor equipment HMI, covering remote diagnosis, source fixes, build verification, rollback backup and acceptance records.
For industrial smoke observation, the delivery covered edge smoke detection, black-smoke color classification, RTSP camera integration, MCP2221A GP0 output linkage, false-positive guards and site verification. Black-smoke classification is not a regulatory emission conclusion.
Based on authorized materials, the delivery included scene screening, 48 simulation images, 4 classes and 40 AI synthetic samples, foreground pose variants and customer review documents for solution discussion and sample-expansion assessment.
The assessment defined vehicle-position unloading, authorized pickup and cargo carry-out events, plus a 2–4 camera layout, Jetson edge analysis, event evidence and acceptance boundaries. It remains a solution-level assessment.
The case covers image capture, OCR recognition, result upload and quality traceability for nameplates, labels, packaging text and equipment numbers, supporting production-line data collection and system integration assessment.
The device combines video capture, image analysis and alarm rules to monitor dust trend changes, supporting local alarms, data logging and remote reporting for site environmental monitoring.
The instrument integrates underwater imaging, lighting, sampling observation, image analysis and data transmission for water ecology monitoring, aquaculture and long-term research observation.
The platform supports multi-protocol device access, edge buffering, rule alarms and platform API synchronization to connect site equipment data with business management systems.
The project covers application migration, driver debugging, system trimming, compatibility verification and deployment documentation for domestic Linux terminal implementation.
The terminal integrates camera, processor, GPS positioning, communication module and night-vision capability for mobile inspection, field supervision, video capture, location records and data upload.