Up to now, researchers have applied segmentation techniques in their studies on teeth images, with construction on tooth root length and depth. In this paper, a new approach to the exact identification of the filled points of the tooth is proposed. In this method, the filled teeth are detection by applying the Bit-Plane algorithm on the OPG images. The novelty of the proposed method is that we can use it in medicine for the detection of dental filling and we calculate and present the area of the filled points which may help dentists to assess the filled point of the tooth. The experimental results, confirmed by the dentists, clearly indicate that this method is able to separate the filled points from the rest of healthy teeth completely.
One of the most considerable challenges in the field of dental health is the detection of defected points which gradually may lead to teeth decay. Dentists can detect and treat dental cavities by viewing X-Ray images, which there exists, the problem that occurs after filling the decayed point which may need detection point in the subsequent photographs of the tooth. Typically, the filled points fail after a period of time and the cavities are revealed again which can cause pain and discomfort in the patient. At this phase, accurate detection of cavities points becomes essential.
In [1], the K-Means algorithm is applied for detection of cavities points on 1 or 2 teeth images not investigated OPG images. In [2] pixel color technique is used to detect the filled points and the gap size between the teeth on the RGB images. Bhan Anupama et al. [3] use the Top-Hat algorithm to identify the tooth cavities on X-Ray images. The advantage of this method is the identification of the cavities, which assists the dentist to identify defected points that do not have an adverse effect on the root of the tooth in a more accurate manner. The segmentation techniques are applied to test root and multitooth length in [4]. In [5], the focus is only on some of the points on the tooth indicate that segmentation is not accurate, and as noted in this article, this method for detecting tooth cavities is inefficient. According to [6], filled points are not precisely identified, while only filled points on adjacent teeth are identified. The run studies, that are so far have identified, are segmented the damages tooth as to detect the root depth with no detection on the whole tooth. In [7], image enhancement is run by watershed and modification of kernel function. This method is not standard in detecting the features of x-ray images. The machine learning classification technique is run by Olsen, Grace F. et al. [8] and the focus is to identify which pixels are classified using the pixel feature vectors. In an experimental result in [9], the focus is on segmentation between both the tooth and alveolar bone and it is important that tooth segmentation methods are presented in the 3D models of tooth and alveolar bones which is required for orthodontic treatment. As demonstrated in [10], the k-means algorithm is run for detecting the cavities and decay tooth. As the results show the root length and cavities are not detected completely, because experimental is through on the surface of the teeth.
According that all points on the OPG images are concern, in this study, for the first time the Bit-Plane algorithm is applied on OPG images to detect the filled point of the tooth and the statistical parameter like area is calculated.
This article is structured as follow: the proposed method is presented in sec.2; the experimental results are expressed in sec.3; and the conclusion is concluded in sec.4.
An algorithm is proposed for the initial segmentation of the images shot to accurately identify points of filled dentures.
For this purpose, first, the obtained images are fed in to Top-Hat algorithm in order to generate Input Original Images, and next, the obtained images are fed into Bit-Plane algorithm in order to generate a target images. By applied Bit-Plane technique the image is divided into eight pages, where changes are made on the 7 th and 8 th pages. The reason for making changes only on these two pages is due to the high value in the logical calculations because they have the most valuable bits in converting pixels of the image in to binary numbers; therefor shifting to valuable bits causes a clear and logical contradiction between pixels and facilitates the objective of separating specific parts of the image, Fig. 1. As it is illustrated in Fig. 1, in the second stage the Top-Hat technique is performed. It can extract small elements and details from Input Original Images that are in previous stage.
In the third stage, it is noticeable to calculating each pixel of given images. These images are seen as a two dimensional matrix and every cell in matrix show a number that can represent pixel value. Because pixel values are decimal number between 0 and 255, each pixel can be represented by an eight bit binary number and all of the bits with different positions make planes of one to eight, Fig. 2 As it is shown in Fig. 3, changing 7 th and 8 th bits on 7 th and 8 th planes can produce a new value for every pixel and this changing is influencing for detecting some points of images that are our target to detection.
Experiments are performed and verified with MATLAB. The filled dentures on the OPG images are well illustrated where parts of the teeth with a mild filling are completely visible through the provided method, Fig. 4. Also the area of filled points of dental are extracted from the Original images and the obtained images after applying our method (in pixels). Based on the parameter the detection of the area of filled points is don
This content is AI-processed based on open access ArXiv data.