Digital Image Enhancement Techniques for Dental Radiographs: A Support to Clinicians

Digital Image Enhancement Techniques for Dental Radiographs: A Support to Clinicians

E. Priya (Sri Sairam Engineering College, India)
Copyright: © 2019 |Pages: 39
DOI: 10.4018/978-1-5225-6243-6.ch001


Dental radiographs suffer frequently from issues such as low contrast and non-uniform illumination. The first step, indeed a significant step, is to enhance these digital images to prepare them for successful post-processing. This pre-processing stage assists to increase the contrast between the foreground that is the teeth and bone from the background regions. In this chapter, image enhancement methods based on spatial, frequency, and spatial-frequency are implemented. The dental radiographs used are available in the public database. The performance of the enhancement methods is validated using qualitative and quantitative measures. It is observed from the results that the enhancement method aids in improving dental features such as crowns, fillings, and bridges. This enables human identification and diagnostic purpose in the way it is possible to identify a variety of diseases. It also prevents the need for a remake, saving the patient from an additional treatment.
Chapter Preview


Human identification is a crucial and prime issue these days as traditional authentication systems may confirm tentative identity. Biometrics is one such technology that is used in identification for uniquely recognizing individuals based on the inherent, physical or behavioural qualities, such as fingerprint, iris, hand vein and voice. Dental biometrics has leading edge when compared to other modalities, which has more complex features and might lead to high error rate (Vijayakumari, Ulaganathan, & Banumathi, 2011; Lin, Lai, & Huang, 2010; Lin, Lai, & Huang, 2012; Rehman, Akram, Faraz, & Riaz, 2015).

Dental biometrics is the primary biometric technique to identify individuals on the basis of their dental characteristics as dental features of human beings are naturally unique. They can be used to authenticate humans precisely or more or less to the highest promising similarity (Rehman, Akram, Faraz, & Riaz, 2015). Thus dental biometrics gets priority over other approaches. Dental biometrics utilizes the evidence revealed by dental radiographs for human identification. This evidence includes tooth contours, relative positions of neighboring teeth, and shapes of the dental features such as crowns, fillings and bridges (Jain, & Chen, 2004).

The conventional biometric characteristics may not be reliable or promising in authentication of humans deceased during natural disasters. In such situations dental features is considered a useful tool with highest accuracy rate (Rehman, Akram, Faraz, & Riaz, 2015). In addition, forensic identification based on dental features relies on the morphology aid in identifying the victims (Zhou, 2010).

Dental radiographs endure frequently from issues such as low contrast and non-uniform illumination that set difficulty to automatically extract the boundaries of teeth for further processing (Rad, Rahim, Rehman, & Saba, 2016; Jain, Chen, & Minut, 2003). Since dental radiographs often suffer from poor quality and uneven exposure, the accuracy of the identification degrades. This complicate the prime stage in further processing such as the task of segmentation, which is the most challenging step (Zhou, 2010). Also for these poor quality images the next vital stage is shape extraction which is an intricate hitch for dental radiographs, where some tooth contours are indiscernible (Jain, & Chen, 2004). Fortunately, the radiographs not only give us the information about the shape of the teeth, but also other information such as the artificial prosthesis of the teeth, the striae patterns and trabecular patterns, etc (Jain, Chen, & Minut, 2003).

The objective of this work is to analyze different image enhancement techniques and to find the most suitable method for dental image radiographs. Image pre-processing is the essential pace in image processing and computer vision, as it makes the processed image more appropriate to exhibit for further analysis. This analysis includes primitive operations such as denoising, contrast enhancement, image smoothing and sharpening, and advanced operations such as image segmentation and others (Adatrao, & Mittal, 2016).

Pre-processing is the first and foremost step in dental image processing and has its importance as the whole processing is based on it. Images of bad quality make difficulties at every stage of feature extraction and matching (Rehman, Akram, Faraz, & Riaz, 2015). This pre-processing stage assists to increase the contrast between the foreground that is the teeth and bony from the background regions. The image enhancement step is performed before segmentation to improve the quality of the dental X-ray images and to separate the teeth in each jaw. This step is crucial as the accuracy of the extracted features depends on the results of image enhancement (Rad, Rahim, Rehman, & Saba, 2016; Zhou, 2010).

Key Terms in this Chapter

Denoising: Denoising is a signal processing method that extract signal from a mixture of signal and noise thus preserving the useful information.

Non-Uniform Illumination: It is the low frequency intensity variations in the image caused by the sources of noise induced by the sensor being used to record the image.

Image Quality Measures: It is a characteristic of an image that measures the perceived image degradation when compared to an ideal or perfect image.

Image Enhancement: Image enhancement improves the quality of images for human perception by removal of noise, reduction of blurring, increase in contrast, and provides more detail information.

Image Pre-Processing: Pre-processing is a common name assigned for the basic operations with images at the lowest level of abstraction which aims at the improvement of image data that suppresses unwanted distortions or enhances some of the image features important for further processing.

Similarity Measures: It quantifies the similarity between two images and shows how self-similar an image is with the query or processed image.

Dental Radiographs: Dental radiographs are x-rays of the teeth, bones, and soft tissues. It helps to find issues related to teeth, mouth, and jaw which cannot be seen during a visual examination.

Histogram: An image histogram is a graphical representation that plots the number of pixels for each intensity value. It is also called as the intensity distribution of the image.

Spatial Domain: Spatial resolution refers to the number of pixels required in construction of a digital image. The images having higher spatial resolution comprise with a greater number of pixels than those of lower spatial resolution.

Complete Chapter List

Search this Book: