Logistics ops · Production

Nanonets automates shipping container recognition for Adani Ports with Computer Vision and OCR

The problem

APSEZ's container inspection was entirely manual, causing misread container numbers due to distance, angle, lighting, and inspector oversight, resulting in delays of up to 10 minutes per container and additional expenses.

Workflow diagram · grounded in source
1
Motion detection at gate
trigger
“Motion detection to monitor entry of containers”
2
Deduplication across cameras
ai_action
“Deduplication of truck at different cameras at a gate”
3
Optimal frame classification
ai_action
“Optimal frame classification”
4
Object Detection and OCR fingerprinting
ai_action
“Object Detection and Optical Character Recognition to fingerprint freight containers”
5
Post-processing fraud verification
validation
“Post processing to verify container details for fraud detection e.g. match number plate captured at the front and back of vehicle, match captured container number at the back, side and top of container”
6
Port system integration
integration
“feeding this information into the port operating system and hence directing the truck accordingly”
7
Continuous model retraining
feedback_loop
“We re-collected the data where the models made an error and built model building processes to redeploy the new models. The system learns over time and accuracy of the system keeps improving.”
Reported outcome

Nanonets deployed an on-premise computer vision and OCR system that fully automated container data capture at port gates, reducing data transfer by over 10,000x and building a self-improving model that increases accuracy over time.

Reported metrics
Container inspection delay (before automation)delays of upto 10 minutes per container
Data transfer reductionover 10,000x
Reported stack
NanonetsOCRDocker
Source
https://nanonets.com/customer-success-story/shipping-container
Read source ↗

Frequently asked questions

What did this team achieve with this AI workflow?

Nanonets deployed an on-premise computer vision and OCR system that fully automated container data capture at port gates, reducing data transfer by over 10,000x and building a self-improving model that increases accur…

What tools did this team use?

Nanonets, OCR, Docker.

What results were reported?

Container inspection delay (before automation): delays of upto 10 minutes per container; Data transfer reduction: over 10,000x (source-reported, not independently verified).

How is this logistics ops AI workflow structured?

Motion detection at gate → Deduplication across cameras → Optimal frame classification → Object Detection and OCR fingerprinting → Post-processing fraud verification → Port system integration → Continuous model retraining.

Related logistics ops cases