Aquamesh line-side automated visual QC for meat in gravy or jelly

Technology
Conceptual
Company

AquaMesh is developing a line-side automated visual QC system for meat-in-gravy and meat-in-jelly products. It converts each sample into objective piece geometry, size distribution, color, meat-to-background fraction, and confidence metrics in under 60 seconds, using calibrated RGB imaging, controlled lighting, cross polarization, and backlighting. Results are logged with SKU, batch, time, and process tags for early drift detection. Replacing a 30-minute manual check with a 1-minute automated update can reduce average post-deviation exposure by about 97% and cut repetitive inspection labor. A six-week pilot is planned to verify accuracy, repeatability, ROI, and production payback using actual line data.

Overview

AquaMesh is developing a line-side automated visual QC cell for packaged meat-in-gravy and meat-in-jelly products. The cell images product at production speed and converts each sample into objective piece geometry, color, and meat-to-background metrics in under 60 seconds. Rather than relying on periodic manual checks and data entry, production teams receive a quantitative status update linked to SKU, batch, time, and process tags.

The main economic opportunity is early detection of visual process drift. On high-speed pouch lines, reducing the inspection interval from 30 minutes to 1 minute can reduce the amount of product made after a process shift from about 2,700 units to about 90 units, a reduction of roughly 96.7%. The system also performs the repetitive inspection and recording work automatically, freeing operators for other tasks.

These economic figures are sensitivity estimates, not guaranteed savings. The validation plan is explicitly designed to replace those assumptions with actual production line data, measured accuracy, rework and giveaway costs, and verified repeatability before commercial deployment is recommended.

Technical specifications

Key features:

  • Calibrated RGB imaging as the production default; VIS/NIR imaging is considered only if blinded testing shows it materially improves accuracy.
  • Controlled diffuse lighting, cross polarization, and backlighting to support consistent measurement on glossy, wet food surfaces.
  • Measured metrics include piece count, size distribution, shape, CIELAB color, projected meat fraction, confidence level, and deviation status.
  • Results are paired with SKU, batch, time, and available process tags so drift can be located at the earliest useful stage.
  • Designed for automated sampling or continuous inspection before filling.
  • Intended to work with high-speed pouch equipment rated at approximately 180, 270, and 360 units per minute.

The proposed validation plan includes 300–500 samples from at least three lots, controlled insertion of defects, reserved blind-test lots, a 10-sample x 3-operator x 3-repeatability study, latency testing, and current-versus-automated workflow timing. A line-side trial then provides verified annual ROI along with a hygienic integration package, production bill of materials, and scale design.

Technology readiness level

This solution is at the proof-of-concept stage. A formal technology readiness rating has not yet been claimed. The proposal outlines a six-week path to validation: mapping current QC practice, selecting a lead SKU and three failure modes, building the calibrated imaging system, collecting and blind-testing representative samples, measuring repeatability and workflow impact, and running a line-side trial.

The production advancement decision depends on meeting agreed accuracy gates and a target production payback of 12 months or less. If the lead SKU passes, a second SKU may be tested. Until the line-side trial is completed, the system should be treated as pre-pilot equipment under development.


About aquamesh

AquaMesh provides an integrated hardware and software stack designed to modernize water quality monitoring through resilient IoT infrastructure. The company’s ecosystem includes the AquaSpectra multi-parameter optical sensor for field deployment, the AquaLab benchtop analyzer, and the AquaLink hub, which bridges sensor traffic to the cloud. This hardware is supported by the AquaView platform, a web-based software suite that uses predictive AI to surface trends, detect anomalies, and generate automated compliance reports. By combining multi-parameter optical sensing with a robust mesh network, AquaMesh offers a more comprehensive data capture solution than traditional single-parameter sondes.

This technology is essential for teams managing critical water systems, including environmental research, stormwater management, and infrastructure monitoring. AquaMesh helps customers shift from manual data collection to continuous, autonomous sensing, reducing the need for on-site reviews through its AI-driven natural language query interface and predictive analytics. The system’s ability to forecast parameter shifts before they reach regulatory thresholds provides operators with actionable insights, improving compliance accuracy and enabling faster responses to water quality changes.

Halo home
Partner smarter. Move faster.
Get new partnering requests
delivered to your inbox.