17-21

Target Strength (TS):

Quantifies the amount of acoustic energy backscattered by a target relative to the incident energy.

Scattering Cross-Section:

Describes how much of the incident energy is scattered by the target.

• Sonar Equation:

  • Scattering models provide estimates of target strength (TSTSTS) for use in sonar performance calculations.

• Object Detection:

  • Distinguishing objects based on size, material, and shape by analyzing backscatter patterns.

• Environmental Studies:

  • Understanding how small particles (e.g., sediment or bubbles) scatter sound to assess underwater conditions.

Rayleigh Scattering

Rayleigh scattering is a phenomenon where sound waves (or light waves) scatter off particles or objects that are much smaller than the wavelength of the incoming wave. It is named after Lord Rayleigh, who first described the phenomenon in the 19th century.

Scattering Regimes depend on kakaka:

• ka≫1: Large spheres; geometric scattering.

• ka≪1: Small spheres; Rayleigh scattering.

• Intermediate values exhibit complex behavior due to creeping waves and material properties.

The scattering from other geometric shapes can be analysed. As with scattering from a sphere, the resulting expressions will often have different forms based on the size of the object relative to the wavelength. For a characteristic dimension l, analytic expressions tend to be valid for either kl ≪ 1 or kl ≫ 1. The domain around kl = 1 is typically complex to analyse.

See page 201

Change detection is the process of comparing multiple images of the same scene taken at different times to identify changes that have occurred over time. In sonar imaging, this technique is crucial for applications such as environmental monitoring, mine hunting, and seafloor mapping.

19.1 Coherent vs Incoherent Change Detection

1. Incoherent Change Detection:

   • Uses only the magnitude data (intensity) of the images.

   • Identifies changes in the reflective properties of the scene.

   • Methods:

     • Differencing: Subtracting intensities pixel by pixel.

     • Ratioing: Dividing intensities pixel by pixel (often displayed in decibels).

2. Coherent Change Detection:

   • Utilizes both the magnitude and phase information of the images.

   • Identifies changes in the structure of the scene.

   • Measures coherence, which quantifies the similarity between the phase patterns of two images.

3. Comparative Advantages:

   • Incoherent:

     • Sensitive to changes in material properties.

     • Robust to noise but less sensitive to structural changes.

   • Coherent:

     • Highly sensitive to structural changes but requires precise alignment (registration).

     • Loss of coherence without a corresponding magnitude change indicates structural deformation.

     • Example: Tracks left by an animal on the seafloor, visible in coherence maps.

Registration ensures that the images are precisely aligned before performing change detection.

1. Importance:

   • Misregistration greater than one resolution cell results in loss of coherence.

   • Even small misalignments can degrade the results, especially in coherent change detection.

2. Accuracy Required:

   • Alignment should be accurate to one-tenth of a resolution cell to minimize coherence loss.

3. Methods:

   • Data-Driven Registration: Comparing and correlating patches of the images to correct positional errors.

   • Marker Points: Using distinct, stable features (e.g., rocky outcroppings) as reference points.

Temporal decorrelation occurs when the coherence between images decreases due to changes in the scene over time.

1. Sources of Temporal Decorrelation:

   • Sediment Transport: Sediment movement caused by waves or currents.

   • Biological Activity: Marine life activity, such as bottom-feeding fish or sand dollars, causing seafloor changes.

   • Environmental Variability: Changes in water properties (e.g., salinity, temperature) affecting sound propagation.

2. Impact:

   • Low coherence across the scene can indicate either:

     • Widespread changes in the scene.

     • Issues with misregistration or other sources of noise.

19.4 Change Detection Methods

1. Feature-Based Techniques

• Focuses on features or objects in the scene rather than pixel-by-pixel comparison.

• Steps:

  • Detect features in each image.

  • Compare the sets of features to identify additions, removals, or changes.

• Applications:

  • Commonly used in mine hunting and regular surveys (e.g., harbors or shipping lanes).

  • Features are cross-referenced with historical databases to identify new or missing objects.

• Advantages:

  • Less dependent on precise registration.

  • Requires prior knowledge about the types of features being detected.

2. Image-Based Techniques

• Compares pixels directly between images.

• Techniques:

  • Differencing

  • Ratioing

  • Both methods may use logarithmic transformations to reduce dynamic range.

• Advantages:

  • Operates directly on the images without requiring prior knowledge.

  • Sensitive to small changes in pixel intensity.

• Limitations:

  • Requires highly accurate registration to ensure reliable results.

Stop-and-Hop Approximation in Sonar Processing

The stop-and-hop approximation is a simplification often used in sonar processing to calculate the travel time of acoustic waves when the sonar platform is in motion. It assumes that the platform is stationary while transmitting and receiving a ping, then "hops" to its next position. This reduces the complexity of the calculations by ignoring the platform's motion during the ping's round-trip.

Image Scaling for Sonar Display

Sonar images often have a high dynamic range, with large differences in backscattered energy between bright points (strong reflections) and dark areas (e.g., shadows). Proper scaling is crucial for displaying these images effectively, as it allows users to discern features across the entire dynamic range.