Abstract:Machine vision and image processing are often used with sensors for situation awareness in autonomous systems, from industrial robots to self-driving cars. The 3D depth sensors, such as LiDAR (Light Detection and Ranging), Radar, are great invention for autonomous systems. Due to the complexity of the setup, LiDAR may not be suitable for some operational environments, for example, a space environment. This study was motivated by a desire to get real-time volumetric and change information with multiple 2D cameras instead of a depth camera. Two cameras were used to measure the dimensions of a rectangular object in real-time. The R-C-P (row-column-pixel) method is developed using image processing and edge detection. In addition to the surface areas, the R-C-P method also detects discontinuous edges or volumes. Lastly, experimental work is presented for illustration of the R-C-P method, which provides the equations for calculating surface area dimensions. Using the equations with given distance information between the object and the camera, the vision system provides the dimensions of actual objects.
What problem does this paper attempt to address?
The main problem that this paper attempts to solve is how to use two 2D cameras to obtain the volume data of an object in real - time without using 3D sensors. Specifically, the author has developed a method named R - C - P (Row - Column - Pixel). Through image processing and edge - detection techniques, it calculates the surface area and volume of rectangular objects from the data captured by two 2D cameras at fixed positions. This method aims to reduce the dependence on expensive and complex 3D sensors, especially in specific operating environments such as space environments, where 3D sensors may be inapplicable due to high power consumption, large amounts of data, complex wiring, high temperature and mechanical requirements, etc.
### Key points of the problem solved by the paper:
1. **Cost - effectiveness**: Compared with 3D sensors, 2D cameras are more cost - effective in installation and operation.
2. **Adaptability**: 2D cameras are suitable for compact or restricted spaces, as well as transparent or highly reflective surfaces, which pose challenges to 3D scanning systems.
3. **Real - time processing ability**: Some systems require immediate data collection, and some 3D scanning techniques may not be able to provide sufficient speed.
4. **Simplified solution**: For applications that do not require precise volume measurement, 2D technology can provide a sufficient approximation.
### Specific implementation of the R - C - P method:
- **Image processing**: First, the RGB images taken from two different angles are converted into grayscale images and then edge - detection is carried out.
- **Edge detection**: The Canny edge - detection algorithm is used to identify the edges in the images.
- **R - C - P algorithm**: Through the detected edge pixels, the R - C - P method can identify and extend the edges and accurately calculate the surface area even if the edges are discontinuous.
- **Volume calculation**: Combining the data from the two cameras, the volume of the object is finally calculated by calculating the width and height detected by each camera.
### Mathematical formulas:
- **Width calculation**:
\[
W1=\left(\frac{l}{p}\right)\times d\quad\text{mm}
\]
\[
W2=\left(\frac{w}{x}\right)\times d\quad\text{mm}
\]
- **Height calculation**:
\[
H1=\left(\frac{l}{q}\right)\times d\quad\text{mm}
\]
\[
H2=\left(\frac{w}{y}\right)\times d\quad\text{mm}
\]
- **Volume calculation**:
\[
V = W1\times\left(\frac{H1 + H2}{2}\right)\times W2\quad\text{mm}^3
\]
### Experimental results:
- The effectiveness of the R - C - P method has been verified through experiments, and the results show that this method can estimate the size of the object well at different distances.
- The experimental data indicate that as the distance between the camera and the object increases, the ratio between the actual size and the calculated size also changes, which provides an important reference for practical applications.
In conclusion, this paper proposes an innovative R - C - P method, which can use 2D cameras to obtain the volume data of an object in real - time without using 3D sensors and has broad application prospects.