**1. Introduction**

With the continuous development of side-scan sonar technology, aspects of side-scan sonars, such as data acquisition stability, sonar image resolution, and image clarity have been improved, providing better technical support for hydrographic surveying and charting. Given the development of marine resource, it is necessary to scan the seabed with side-scan sonars to grasp the general information of the seabed scene and topography for many applications, such as offshore oil drilling, channel dredging, submarine pipeline detection, seabed structure detection, marine environment detection, marine archaeology, and detection and location of large-scale seabed targets [1–4].

In Figure 1, a side-scan sonar is scanning the seabed scene. The side-scan sonar transducer installed on both sides of the Autonomous Underwater Vehicle (AUV) emits spherical acoustic signals. After reflection from the seabed, the reflected signals are collected and received by the receiver according to the transmission time of the sonar signal. A side-scan sonar image is formed by converting the reflected signal intensity into the gray level. However, the energy of the sonar acoustic wave can attenuate in water. There are three main types of attenuation. The first category is physical attenuation, the second is the absorption of seawater and the third is echo attenuation [5–7]. In addition to the causes of sonar energy attenuation, side-scan sonar image is also affected by beam patterns, angular responses of different sediments, and changes in seabed topography [7–9]. Based on the above reasons, the gray-scale of side-scan sonar images is uneven.

**Figure 1.** Working schematic diagram of a side-scan sonar.

As shown in Figure 2, the original side-scan sonar image has uneven gray distribution, which affects the interpretation of the side-scan sonar image and the subsequent image processing. Therefore, gray scale correction should be conducted before processing the side-scan sonar image, such as image matching, stitching, and target recognition [10–12].

**Figure 2.** Original side-scan sonar image.

#### **2. Current Gray Scale Correction Methods for Side-Scan Sonar Images**

At present, many kinds of gray scale correction methods are available for side-scan sonar images, which can be categorized into six kinds of methods.

#### *2.1. Time Variant Gain (TVG)*

The TVG method is a commonly used method for gray scale correction of side-scan sonar images. The time variant gain method, adopted by Johnson et al., is based on the distance between each point of the seabed and the sonar array in the side-scan sonar image [9,13]. The side-scan sonar images are compensated using Equation (1):

$$\text{EL} = 2\text{TL} - \text{TS} = 30\text{lgR} + 2a\text{R}/10^3 - \text{S}\_f \tag{1}$$

where *EL* is the compensation amount, *TL* is the energy loss caused by the propagation process, *TS* is the target strength, *R* is the propagation distance, α is the absorption coefficient, *Sf* is the seabed backscattering intensity. Since α and *R* cannot be easily obtained, their empirical values are required in the compensation. The TVG method is usually implemented using hardware. While the intensity can be compensated for, to a certain extent, it is impossible to mimic the same sonar energy attenuation process. Sometimes unrealistic gain parameters cause secondary gray distortion. Two problems arise if we use the algorithm in Equation (1) to gray correct side-scan sonar images. (1) It is difficult to determine the specific values of the parameters in Equation (1) only using side-scan sonar images, so imbalanced correction may occur. (2) Different side-scan sonar images require different parameters in the TVG equation to achieve better image enhancement, so the algorithm is not universal.

#### *2.2. Histogram Equalization (HE)*

Histogram equalization is used to improve the uniformity of the gray distribution of the side-scan sonar image by adjusting the gray distribution of the entire image. The essence of histogram equalization involves enlarging the gray level difference of the image, and the overall contrast of the image is improved after equalization. HE has been widely used because of its simplicity and directness [14]. However, HE often increases image noise, and because some gray scale merging results in blurring the weak edges of the image, it leads to over-enhancement in the regions with large histogram peaks, which is extremely disadvantageous to side-scan sonar image processing. Therefore, histogram equalization is not the best method to address gray distortion in side-scan sonar images.
