Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

LiqD: A Dynamic Liquid Level Detection Model under Tricky Small Containers

Yukun Ma, Zikun Mao

TL;DR

This work tackles dynamic liquid level detection in small containers using non-contact image analysis. It introduces a data-efficient pipeline that leverages the Segment Anything Model (SAM) for data generation, the SemiReward framework for pseudo-label filtering, and the U²-Net for precise container masking, followed by morphological refinement and grayscale frame differencing to isolate liquid-level changes. A lightweight classifier operates on binary delta maps to determine liquid-level states, enabling robust detection with limited labeled data. The approach demonstrates superior accuracy and low error across comparisons to several baselines, offering a practical, generalizable solution for real-world monitoring in industrial and domestic contexts.

Abstract

In daily life and industrial production, it is crucial to accurately detect changes in liquid level in containers. Traditional contact measurement methods have some limitations, while emerging non-contact image processing technology shows good application prospects. This paper proposes a container dynamic liquid level detection model based on U^2-Net. This model uses the SAM model to generate an initial data set, and then evaluates and filters out high-quality pseudo-label images through the SemiReward framework to build an exclusive data set. The model uses U^2-Net to extract mask images of containers from the data set, and uses morphological processing to compensate for mask defects. Subsequently, the model calculates the grayscale difference between adjacent video frame images at the same position, segments the liquid level change area by setting a difference threshold, and finally uses a lightweight neural network to classify the liquid level state. This approach not only mitigates the impact of intricate surroundings, but also reduces the demand for training data, showing strong robustness and versatility. A large number of experimental results show that the proposed model can effectively detect the dynamic liquid level changes of the liquid in the container, providing a novel and efficient solution for related fields.

LiqD: A Dynamic Liquid Level Detection Model under Tricky Small Containers

TL;DR

This work tackles dynamic liquid level detection in small containers using non-contact image analysis. It introduces a data-efficient pipeline that leverages the Segment Anything Model (SAM) for data generation, the SemiReward framework for pseudo-label filtering, and the U²-Net for precise container masking, followed by morphological refinement and grayscale frame differencing to isolate liquid-level changes. A lightweight classifier operates on binary delta maps to determine liquid-level states, enabling robust detection with limited labeled data. The approach demonstrates superior accuracy and low error across comparisons to several baselines, offering a practical, generalizable solution for real-world monitoring in industrial and domestic contexts.

Abstract

In daily life and industrial production, it is crucial to accurately detect changes in liquid level in containers. Traditional contact measurement methods have some limitations, while emerging non-contact image processing technology shows good application prospects. This paper proposes a container dynamic liquid level detection model based on U^2-Net. This model uses the SAM model to generate an initial data set, and then evaluates and filters out high-quality pseudo-label images through the SemiReward framework to build an exclusive data set. The model uses U^2-Net to extract mask images of containers from the data set, and uses morphological processing to compensate for mask defects. Subsequently, the model calculates the grayscale difference between adjacent video frame images at the same position, segments the liquid level change area by setting a difference threshold, and finally uses a lightweight neural network to classify the liquid level state. This approach not only mitigates the impact of intricate surroundings, but also reduces the demand for training data, showing strong robustness and versatility. A large number of experimental results show that the proposed model can effectively detect the dynamic liquid level changes of the liquid in the container, providing a novel and efficient solution for related fields.
Paper Structure (17 sections, 7 equations, 7 figures, 1 table)

This paper contains 17 sections, 7 equations, 7 figures, 1 table.

Table of Contents

  1. Introduction
  2. Related Works
  3. SAM Model
  4. U²-Net
  5. Bottleneck in Hand-crafted Design
  6. Morphological Compensation
  7. Grayscale Value Conversion
  8. Method
  9. Construct Data Engine
  10. Salient Target Extraction
  11. Container Morphology Compensation
  12. Dynamic Liquid Level Detection
  13. Threshold Division
  14. Liquid level difference calculation
  15. Liquid level detection
  16. ...and 2 more sections

Figures (7)

  • Figure 1: Overall framework
  • Figure 2: Before the Completion
  • Figure 3: After the Completion
  • Figure 4: Threshold Division
  • Figure 5: Threshold Data
  • ...and 2 more figures