SaeroSense새로센스
  1. Home
  2. Services
  3. Images & video

Service · Images & video

Images & video: computer vision and visual inspection

We build computer-vision pipelines for inspection and monitoring: detection and segmentation models trained on your imagery, deployed as a service your line or field team can actually use.

Selected work

Vision systems, from bridge piers to MRI scans.

Real outputs from projects our founding team has delivered. Open any project for the workflow, the models and the results.

Inspection & industrial4 projects

Concrete defect segmentation: project visualClient project

Infrastructure inspection · South Korea

Concrete defect segmentation

One model outlines five kinds of concrete damage, so inspectors measure defects instead of eyeballing photos.

Result: 5 damage classes · 5 models benchmarked

View case study
PPE compliance detection: project visualApplied R&D

Industrial safety

PPE compliance detection

Site cameras check seven types of protective equipment automatically, so safety teams don’t have to watch every feed.

Result: 7 PPE classes · 2.3k training images

View case study
Blurry license plate recognition: project visualApplied R&D

Smart city · traffic & parking

Blurry license plate recognition

Reads plates from the low-quality street CCTV operators already have, with no camera upgrade needed.

Result: >90% recognition accuracy

View case study
Applied R&D
94.8%AUC-ROC · genuine vs counterfeit

Retail · brand protection

Luxury logo authentication

Automated first-line screening of genuine vs. counterfeit logos for brand-protection teams and resale platforms.

Result: 94.8% AUC-ROC · 94.0% F1 (brand A)

View case study

Autonomy & localization2 projects

Environment perception for self-driving: project visualApplied R&D

Autonomous driving

Environment perception for self-driving

Turns a camera’s segmentation map into what a vehicle planner needs first: where it can drive and where the lanes are.

Result: Drivable space + lanes, one model

View case study
Stereo visual-inertial odometry (MSCKF): project visualApplied R&D

Robotics · drones · AR/VR

Stereo visual-inertial odometry (MSCKF)

Accurate positioning without GPS for robots, drones and AR devices: under 0.2 m error over the evaluated run.

Result: <0.2 m position error

View case study

Medical imaging2 projects

Brain tumor segmentation in MRI: project visualMedical AI research

Medical imaging

Brain tumor segmentation in MRI

Consistent, automatic tumor outlines across whole MRI volumes: a step toward faster review and less variation.

Result: 3D U-Net · multi-modal MRI

View case study
Generative AI for knee X-rays: project visualHospital research

Medical imaging · radiology

Generative AI for knee X-rays

With a university hospital: can generative models produce realistic knee X-rays to improve image quality and data?

Result: 3 generative model families compared

View case study

Client names are withheld unless public. References are available on request.

All case studies

What this covers

Built end to end, not just the model.

  • Defect detection & segmentation

    Surface defects, structural flaws, assembly errors; pixel-level segmentation when the shape and extent of a defect matters, not just its presence.

  • Dataset strategy

    Labeling plans, augmentation, and handling the class imbalance that dominates real inspection data, where defects are rare by definition.

  • Edge and server inference

    Models sized for your constraint: GPU server, industrial PC, or embedded device.

  • Human-in-the-loop

    Review queues and confidence thresholds so the system earns trust instead of demanding it.

Where it works well

Manufacturing QC, infrastructure and construction inspection (cracks, spalling, corrosion), and monitoring tasks where a person currently reviews images one by one.

Common questions

Before you ask.

Q01How many labeled images do we need to start?

Often fewer than teams expect. With modern pretrained backbones and careful augmentation, useful prototypes frequently start from hundreds (not tens of thousands) of labeled examples, and the labeling plan is part of the engagement.

Q02Can models run on our existing hardware?

Usually yes. We profile your target (GPU server, industrial PC, or edge device) and choose or distill models to fit its latency and memory budget.

Q03What accuracy can we expect?

We never promise a number before seeing data. Scoping includes a quick feasibility pass on your imagery so expectations are set from evidence, not sales copy.

Start a project

Discuss your project.

Send a few lines about your data and the outcome you want. We reply within 1 business day with a feasibility read and a proposed scoping plan.

Send us a message contact@saerosense.com Seoul based · available throughout US business hours