HACARUS Inspect for Noise InspectionHACARUS Inspect for Noise Inspection

Detecting abnormalities with smarter analysis - suitable for all environments.

Noise inspection is used in various situations, such as industrial sites for machinery inspections, bridges and tunnels for infrastructure defect detection, and factories for product inspections.
Due to a lack of technicians with a “trained ear”, and an abundance of inspection sites, noise inspection is currently an under-used approach.
Based in Sparse Modeling AI, HACARUS’ noise inspection services can be applied in various environments with ease - and with less training data.
Our model’s lightweight design allows for greater flexibility, enabling integration into existing hardware without the need for an external network connection.
We provide constant inspection capabilities for factories as well applications for bridge and tunnel monitoring using sound collection microphones.

  • Ease of integration in existing
    systems with lightweight AI

  • Accurate results,
    regardless of the environment

  • Monitoring capabilities without
    network connection

Service Specifications

Product type: API or SDK
Development language: C++ or Python (embedded development in existing terminals available)
Operating system: Windows or Linux (GPU not required)

Proven Superior Performance

SPECTRO(SDK) & Macnica SENSPIDER

200 times faster

Utilizing SENSPIDER, MACNICA’s AI-ready IoT sensor hub combined with SPECTRO CORE’s versatile algorithms, the joint solution can detect anomalies in vibration data. In the case study highlighted on the right, using data from industrial fans, the solution can detect anomalies over 200 times faster compared with a common k-Nearest Neighbors (KNN) approach, with 100% accuracy.

KNN HACARUS Inspect
Prediction Time 212.876 µs 973 µs
Training Time 33 sec. 14.36 sec.
Accuracy 100% 100%
Precision 100% 100%
Recall 100% 100%
MACNICA

Provides AI analysis of frequency data and time series data/h3>
Time series data (RAW data)

Time series data (RAW data)

Frequency spectrum (FFT)

Frequency spectrum (FFT)

Use Case Examples

Defect Detection in Factory Machinery

Defect Detection in Factory Machinery

High speed detection for noise abnormalities related to machines by constantly collecting chronological data.

Defect Detection in Bridges and Tunnels

Defect Detection in Bridges and Tunnels

Providing highly accurate defect detection for individualized environments in bridges and tunnels from sound collection of strikes within the structure.

Predictive Maintenance & Component Health

Predictive Maintenance & Component Health

Capable of analyzing time-series data, both in the frequency and time domains, such as sound and vibrations produced by machinery to provide insights about component health, enabling preemptive maintenance.

Example of use: Ensure continuous operation, predict component failure.

Subscribe to our newsletter

Click here to sign up