TWI749909B - Image analysis system and image analysis method - Google Patents

Image analysis system and image analysis method Download PDF

Info

Publication number
TWI749909B
TWI749909B TW109141592A TW109141592A TWI749909B TW I749909 B TWI749909 B TW I749909B TW 109141592 A TW109141592 A TW 109141592A TW 109141592 A TW109141592 A TW 109141592A TW I749909 B TWI749909 B TW I749909B
Authority
TW
Taiwan
Prior art keywords
image
tooth
plaque
ultraviolet light
white light
Prior art date
Application number
TW109141592A
Other languages
Chinese (zh)
Other versions
TW202221647A (en
Inventor
林庚達
胡博期
潘柏瑋
楊昇宏
Original Assignee
財團法人金屬工業研究發展中心
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 財團法人金屬工業研究發展中心 filed Critical 財團法人金屬工業研究發展中心
Priority to TW109141592A priority Critical patent/TWI749909B/en
Application granted granted Critical
Publication of TWI749909B publication Critical patent/TWI749909B/en
Publication of TW202221647A publication Critical patent/TW202221647A/en

Links

Images

Landscapes

  • Dental Tools And Instruments Or Auxiliary Dental Instruments (AREA)

Abstract

An image analysis system and an image analysis method are provided. An image analysis system includes a storage device, an image capture device and a processing circuit. The storage device stores a tooth-position identify module and a dental plaque identify module. The image capture module captures white light image and UV light image corresponding to teeth. The processing circuit inputs the white light image into the tooth-position identify module and the tooth-position identify module outputs a teeth image with multiple teeth areas corresponding to teeth of multiple teeth positions. The processing circuit inputs the UV light image into the dental plaque identify module and the dental plaque identify module outputs a dental plaque distribution image according the UV light image. The processing circuit analyses each dental plaque distribution of the teeth areas and creates a dental plaque value table.

Description

影像分析系統與影像分析方法Image analysis system and image analysis method

本發明是有關於一種影像分析,且特別是有關於一種影像分析系統與影像分析方法。The present invention relates to an image analysis, and particularly relates to an image analysis system and an image analysis method.

牙菌斑數量是衡量一個人的牙齒健康程度的一個參考指標。傳統上,牙菌斑的檢測是經由醫師透過肉眼逐顆牙齒比對與檢查,並以人工的方式將每一顆牙齒的牙菌斑分布情形登記到表格上。然而,此種紀錄方式除了效率不佳外,更需要消耗醫療人員的時間與精神。The number of dental plaque is a reference index to measure a person's dental health. Traditionally, the detection of dental plaque is performed by a physician comparing and examining each tooth with the naked eye, and manually registering the plaque distribution of each tooth on a form. However, this recording method is not only inefficient, but also consumes the time and energy of medical staff.

本發明提供一種影像分析系統與影像分析方法,可自動判斷齒號以及牙菌斑分布情形。The invention provides an image analysis system and an image analysis method, which can automatically determine the tooth number and the distribution of dental plaque.

本發明的影像分析系統包括儲存裝置、取像裝置以及處理電路。儲存裝置用以儲存牙位辨識模組以及牙菌斑辨識模組。取像裝置用以取得對應於牙齒的白光影像以及紫外光影像。處理電路耦接儲存裝置以及取像裝置。在本實施例中,處理電路將白光影像輸入至牙位辨識模組,並且牙位辨識模組輸出被定義有對應於牙齒的多個牙位的多個牙齒區域的牙齒影像。在本實施例中,處理電路將紫外光影像輸入至牙菌斑辨識模組,並且牙菌斑辨識模組依據紫外光影像來輸出牙菌斑分布影像。在本實施例中,處理電路藉由牙位辨識模組比對牙菌斑分布影像以及牙齒影像,以分析牙齒區域中各別的牙菌斑分布,並且建立對應於牙齒的牙菌斑數值表。The image analysis system of the present invention includes a storage device, an image capturing device and a processing circuit. The storage device is used for storing the tooth position identification module and the dental plaque identification module. The image capturing device is used to obtain white light images and ultraviolet light images corresponding to the teeth. The processing circuit is coupled to the storage device and the image capturing device. In this embodiment, the processing circuit inputs the white light image to the tooth position recognition module, and the tooth position recognition module outputs tooth images defined with a plurality of tooth regions corresponding to a plurality of tooth positions of the tooth. In this embodiment, the processing circuit inputs the ultraviolet light image to the dental plaque recognition module, and the dental plaque recognition module outputs the plaque distribution image according to the ultraviolet light image. In this embodiment, the processing circuit compares the plaque distribution image and the tooth image with the tooth position recognition module to analyze the individual plaque distribution in the tooth region, and establish a plaque value table corresponding to the tooth .

本發明的影像分析方法包括以下步驟:透過取像裝置取得對應於牙齒的白光影像以及紫外光影像;將白光影像輸入至牙位辨識模組,以使牙位辨識模組輸出被定義有對應於牙齒的多個牙位的多個牙齒區域的牙齒影像;將紫外光影像輸入至牙菌斑辨識模組,以使牙菌斑辨識模組依據紫外光影像來輸出牙菌斑分布影像;以及藉由牙位辨識模組比對牙菌斑分布影像以及牙齒影像,以分析些牙齒區域中各別的牙菌斑分布,並且建立對應於牙齒的牙菌斑數值表。The image analysis method of the present invention includes the following steps: obtaining white light images and ultraviolet light images corresponding to the teeth through the image capturing device; Tooth images of multiple tooth regions and multiple tooth positions of the tooth; input the ultraviolet light image to the plaque recognition module so that the plaque recognition module outputs the plaque distribution image according to the ultraviolet light image; and The tooth position recognition module compares the plaque distribution image and the tooth image to analyze the individual plaque distribution in some tooth regions, and establish a plaque value table corresponding to the teeth.

基於上述,本發明的影像分析系統與影像分析方法可自動判斷齒號以及牙菌斑分布情形,有效地減少檢查牙菌斑所需的時間。Based on the above, the image analysis system and the image analysis method of the present invention can automatically determine the tooth number and the distribution of dental plaque, effectively reducing the time required for the inspection of dental plaque.

為讓本發明的上述特徵和優點能更明顯易懂,下文特舉實施例,並配合所附圖式作詳細說明如下。In order to make the above-mentioned features and advantages of the present invention more comprehensible, the following specific embodiments are described in detail in conjunction with the accompanying drawings.

為了使本發明之內容可以被更容易明瞭,以下特舉實施例作為本發明確實能夠據以實施的範例。另外,凡可能之處,在圖式及實施方式中使用相同標號的元件/構件/步驟,是代表相同或類似部件。In order to make the content of the present invention more comprehensible, the following embodiments are specifically cited as examples on which the present invention can indeed be implemented. In addition, wherever possible, elements/components/steps with the same reference numbers in the drawings and embodiments represent the same or similar parts.

並且,除非另有定義,本文使用的所有術語(包括技術和科學術語)具有與本發明所屬領域的普通技術人員通常理解的相同的含義。將進一步理解的是,諸如在通常使用的字典中定義的那些術語應當被解釋為具有與它們在相關技術和本發明的上下文中的含義一致的含義,並且將不被解釋為理想化的或過度正式的意義,除非本文中明確地這樣定義。And, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which the present invention belongs. It will be further understood that terms such as those defined in commonly used dictionaries should be interpreted as having meanings consistent with their meaning in the context of related technologies and the present invention, and will not be interpreted as idealized or excessive The formal meaning, unless explicitly defined as such in this article.

圖1是依照本發明的一實施例的一種影像分析系統的示意圖。參考圖1,在本實施例中,影像分析系統100包括儲存裝置150、取像裝置120以及處理電路130。儲存裝置150用以儲存牙位辨識模組151以及牙菌斑辨識模組152。取像裝置120用以取得對應於牙齒的白光影像以及紫外光影像。處理電路130耦接儲存裝置150以及取像裝置120。在本實施例中,處理電路130將白光影像輸入至牙位辨識模組151,並且牙位辨識模組151輸出被定義有對應於牙齒的多個牙位的多個牙齒區域的牙齒影像。並且,處理電路130將紫外光影像輸入至牙菌斑辨識模組152,並且牙菌斑辨識模組依據紫外光影像來輸出牙菌斑分布影像。如此一來,處理電路130可藉由牙位辨識模組比對牙菌斑分布影像以及牙齒影像,以分析所述多個牙齒區域中各別的牙菌斑分布,並且建立對應於牙齒的牙菌斑數值表。因此,本發明的影像分析系統可自動判斷齒號以及牙菌斑分布情形,有效地減少檢查牙菌斑所需的時間。FIG. 1 is a schematic diagram of an image analysis system according to an embodiment of the invention. 1, in this embodiment, the image analysis system 100 includes a storage device 150, an image capturing device 120 and a processing circuit 130. The storage device 150 is used to store the tooth position recognition module 151 and the dental plaque recognition module 152. The image capturing device 120 is used to obtain white light images and ultraviolet light images corresponding to the teeth. The processing circuit 130 is coupled to the storage device 150 and the image capturing device 120. In this embodiment, the processing circuit 130 inputs the white light image to the tooth position recognition module 151, and the tooth position recognition module 151 outputs tooth images defined with multiple tooth regions corresponding to multiple teeth positions of the tooth. In addition, the processing circuit 130 inputs the ultraviolet light image to the dental plaque recognition module 152, and the dental plaque recognition module outputs the plaque distribution image according to the ultraviolet light image. In this way, the processing circuit 130 can compare the plaque distribution image and the tooth image with the tooth position recognition module to analyze the respective plaque distribution in the multiple tooth regions, and create a tooth corresponding to the tooth. Plaque value table. Therefore, the image analysis system of the present invention can automatically determine the tooth number and the distribution of dental plaque, effectively reducing the time required for the inspection of dental plaque.

在本實施例中,儲存裝置150可例如包括隨機存取記憶體(Random-Access Memory;RAM)、唯讀記憶體(Read-Only Memory;ROM)、光碟(Optical disc)、磁碟(Magnetic disk)、硬驅動機(Hard drive)、固態驅動機(Solid-state drive)、快閃驅動機(Flash drive)、安全數位(Security digital;SD)卡、記憶條(Memory stick)、緊密快閃(Compact flash;CF)卡或任何類型的儲存設備,但本發明不以此為限。在本實施例中,取像裝置120可例如包括可見光攝影機、紅外線(Infrared light)攝影機、熱顯像攝影機、彩色影像深度攝影機,但本發明並不以此為限。在本實施例中,處理電路130可例如包括中央處理器(Central Processing Unit,CPU)、微處理器(Microprocessor Control Unit,MCU)或現場可程式閘陣列(Field Programmable Gate Array,FPGA),但本發明並不以此為限。In this embodiment, the storage device 150 may include, for example, random-access memory (Random-Access Memory; RAM), read-only memory (Read-Only Memory; ROM), optical disc (Optical disc), and magnetic disk (Magnetic disk). ), Hard drive, Solid-state drive, Flash drive, Security digital (SD) card, Memory stick, compact flash ( Compact flash; CF) card or any type of storage device, but the present invention is not limited to this. In this embodiment, the image capturing device 120 may include, for example, a visible light camera, an infrared (Infrared light) camera, a thermal imaging camera, and a color image depth camera, but the invention is not limited to this. In this embodiment, the processing circuit 130 may include, for example, a central processing unit (CPU), a microprocessor (Microprocessor Control Unit, MCU), or a field programmable gate array (Field Programmable Gate Array, FPGA). The invention is not limited to this.

圖2是依照本發明的一實施例的影像分析方法的流程圖。參考圖1及圖2,影像分析系統100可執行以下步驟S210~S240,以建立對應於牙齒的牙菌斑數值表。在步驟S210中,影像分析系統100可經由白光光源101照明牙齒。並且,當白光光源101照明牙齒時,影像分析系統100可透過取像裝置120拍攝牙齒,以取得白光影像。在步驟S220中,影像分析系統100可經由紫外光光源102照明牙齒。並且,當紫外光光源102照明牙齒時,影像分析系統100可透過取像裝置120拍攝牙齒,以取得紫外光影像。FIG. 2 is a flowchart of an image analysis method according to an embodiment of the invention. 1 and 2, the image analysis system 100 may perform the following steps S210 to S240 to create a table of plaque values corresponding to teeth. In step S210, the image analysis system 100 can illuminate the teeth through the white light source 101. Moreover, when the white light source 101 illuminates the teeth, the image analysis system 100 can capture the teeth through the image capturing device 120 to obtain white light images. In step S220, the image analysis system 100 can illuminate the teeth via the ultraviolet light source 102. In addition, when the ultraviolet light source 102 illuminates the teeth, the image analysis system 100 can photograph the teeth through the image capturing device 120 to obtain an ultraviolet image.

需要說明的是,參考圖3A至圖3E,在步驟S210以及步驟S220中,取像裝置120可對應於如圖3A至圖3E的不同的取樣位置,拍攝多組白光影像以及紫外光影像,但本發明不以此為限。圖3A是依照本發明的一實施例的牙齒的頰側的正面影像310的示意圖。圖3B是依照本發明的一實施例的牙齒的頰側的一側面影像320的示意圖。圖3C是依照本發明的一實施例的牙齒的頰側的另一側面影像330的示意圖。圖3D是依照本發明的一實施例的牙齒的咬合側的上顎影像340的示意圖。圖3E是依照本發明的一實施例的牙齒的咬合側的下顎影像350的示意圖。也就是說,取像裝置120可在不同時間從對應於牙齒的同一取樣位置來取得白光影像以及紫外光影像。並且,取像裝置120可從對應於牙齒的多個不同取樣位置,取得多組如圖3A至圖3E的白光影像以及紫外光影像。It should be noted that, referring to FIGS. 3A to 3E, in step S210 and step S220, the imaging device 120 may correspond to different sampling positions as shown in FIGS. 3A to 3E, and shoot multiple sets of white light images and ultraviolet light images, but The present invention is not limited to this. 3A is a schematic diagram of a frontal image 310 of the buccal side of a tooth according to an embodiment of the present invention. FIG. 3B is a schematic diagram of a silhouette image 320 of the buccal side of a tooth according to an embodiment of the present invention. FIG. 3C is a schematic diagram of another silhouette image 330 of the buccal side of a tooth according to an embodiment of the present invention. 3D is a schematic diagram of an upper jaw image 340 on the occlusal side of a tooth according to an embodiment of the present invention. 3E is a schematic diagram of a lower jaw image 350 on the occlusal side of a tooth according to an embodiment of the present invention. In other words, the imaging device 120 can obtain white light images and ultraviolet light images from the same sampling position corresponding to the tooth at different times. In addition, the image capturing device 120 can obtain multiple sets of white light images and ultraviolet light images as shown in FIGS. 3A to 3E from multiple different sampling positions corresponding to the teeth.

附帶一提,在本實施例中,影像分析系統100可包括光源驅動電路110,用以驅動白光光源101以及紫外光光源102。舉例而言,在步驟S210中,光源驅動電路110可驅動白光光源101,使白光光源101照明牙齒。並且,在步驟S220中,光源驅動電路110可驅動紫外光光源102,使紫外光光源102照明牙齒。經由光源驅動電路110驅動不同的光源,取像裝置120可對應地分別取得白光影像以及紫外光影像。值得注意的是,在本實施例中,紫外光光源102的亮度值可大於白光光源101的亮度值,以便後續的影像處理有較好的效果,但本發明並不限於此。在其他實施例中,白光光源101的亮度值可大於紫外光光源102的亮度值。或者,取像裝置120取得紫外光影像的曝光時間長度可大於取得白光影像的曝光時間長度,取得類似於紫外光光源102的亮度值大於白光光源101的亮度值的效果。又或者,取像裝置120可將取得紫外光影像的白平衡設定不同於取得白光影像的白平衡設定,取得類似於紫外光光源102的亮度值大於白光光源101的亮度值的效果。Incidentally, in this embodiment, the image analysis system 100 may include a light source driving circuit 110 for driving the white light source 101 and the ultraviolet light source 102. For example, in step S210, the light source driving circuit 110 can drive the white light source 101 so that the white light source 101 illuminates the teeth. In addition, in step S220, the light source driving circuit 110 can drive the ultraviolet light source 102 so that the ultraviolet light source 102 illuminates the teeth. By driving different light sources through the light source driving circuit 110, the image capturing device 120 can obtain the white light image and the ultraviolet light image respectively. It is worth noting that in this embodiment, the brightness value of the ultraviolet light source 102 may be greater than the brightness value of the white light source 101, so that the subsequent image processing has a better effect, but the present invention is not limited to this. In other embodiments, the brightness value of the white light source 101 may be greater than the brightness value of the ultraviolet light source 102. Alternatively, the length of the exposure time for the image capturing device 120 to obtain the ultraviolet image may be greater than the length of the exposure time for the white light image to obtain an effect similar to that the brightness value of the ultraviolet light source 102 is greater than the brightness value of the white light source 101. Alternatively, the image capturing device 120 may set the white balance setting for obtaining the ultraviolet light image different from the white balance setting for obtaining the white light image, so as to obtain an effect similar to that the brightness value of the ultraviolet light source 102 is greater than the brightness value of the white light source 101.

回到圖2,在步驟S211中,參考圖3A至圖4,處理電路130可將如圖3A至圖3E的白光影像輸入至牙位辨識模組151,以使牙位辨識模組151輸出被定義有對應於牙齒的多個牙位的多個牙齒區域401-1~401-N(N為大於1的正整數)的牙齒影像400。具體而言,處理電路130可將如圖3A至圖3E的白光影像輸入至牙位辨識模組151。並且,處理電路130可依據牙齒資料庫對牙齒影像400進行牙位標記,以藉由影像辨識牙齒影像400中的多個牙齒特徵來定義在牙齒影像400中的所述多個牙位的所述多個牙齒區域401-1~401-N。舉例而言,處理電路130可使用faster-RCNN、YoLo等模型建立深度學習物件偵測演算法,並採用特徵抽取方法VGG16或Resnet101作為骨幹網路(backbone network),以完成牙齒推論模型。接著,處理電路130可依據牙齒推論模型框選出白光影像中每顆牙齒的位置,以輸出被定義有對應於牙齒的多個牙位的多個牙齒區401-1~401-N的牙齒影像400。Returning to FIG. 2, in step S211, referring to FIGS. 3A to 4, the processing circuit 130 can input the white light image as shown in FIGS. 3A to 3E to the tooth position recognition module 151, so that the tooth position recognition module 151 can output A tooth image 400 of a plurality of tooth regions 401-1 to 401-N (N is a positive integer greater than 1) corresponding to a plurality of tooth positions of the tooth is defined. Specifically, the processing circuit 130 can input the white light images shown in FIGS. 3A to 3E to the tooth position recognition module 151. In addition, the processing circuit 130 may mark the tooth position of the tooth image 400 according to the tooth database, so as to identify a plurality of tooth features in the tooth image 400 to define the plurality of tooth positions in the tooth image 400 Multiple tooth areas 401-1~401-N. For example, the processing circuit 130 can use models such as faster-RCNN and YoLo to establish a deep learning object detection algorithm, and use the feature extraction method VGG16 or Resnet101 as a backbone network to complete the tooth inference model. Then, the processing circuit 130 can select the position of each tooth in the white light image according to the tooth inference model frame to output a tooth image 400 defined with a plurality of tooth regions 401-1 to 401-N corresponding to a plurality of tooth positions of the tooth .

在步驟S221中,參考圖5A至圖5C,處理電路130可將紫外光影像510輸入至牙菌斑辨識模組152,以使牙菌斑辨識模組152依據紫外光影像510輸出牙菌斑分布影像520。具體而言,牙菌斑辨識模組152可將紫外光影像510轉換為HSV格式,並使用螢光範圍閥值(或稱預設範圍)來找出紫外光影像510中的牙菌斑的位置。舉例而言,原始的紫外光影像510可為RGB色彩空間的格式。並且,牙菌斑辨識模組152可將紫外光影像510從RGB色彩空間轉換至HSV色彩空間。再者,牙菌斑辨識模組152可依據預設範圍來判斷經色彩空間轉換後的紫外光影像510中的多個牙菌斑像素521,以產生如圖5B的牙菌斑分布影像520。在本實施例中,圖5B的牙菌斑分布影像520可為二值化影像。接著,牙菌斑辨識模組152可將圖5B的牙菌斑分布影像520以及圖5A紫外光影像510進行影像堆疊,以產生如圖5C的包含有多個牙菌斑像素531的堆疊影像530,但本發明並不限於此。在一實施例中,牙菌斑辨識模組152可將圖5B的牙菌斑分布影像520以及類似於圖4的牙齒影像400(白光影像)進行影像堆疊,以產生以白光影像的牙齒影像為背景的堆疊影像,以使同樣可用於表示牙菌斑分布資訊。影像分析系統100還可透過顯示裝置140顯示堆疊影像530,以提供牙菌斑分布資訊予醫療人員進行判斷。另外,在其他實施例中,牙菌斑辨識模組152可將紫外光影像510從第一色彩空間的格式轉換至第二色彩空間的格式,其中所述第一色彩空間以及所述第二色彩空間可各別任選自RGB色彩空間、HSV色彩空間或HSL色彩空間等各種不同的色彩空間,而不限於上述實施方式。In step S221, referring to FIGS. 5A to 5C, the processing circuit 130 may input the ultraviolet light image 510 to the dental plaque recognition module 152, so that the dental plaque recognition module 152 outputs the plaque distribution according to the ultraviolet light image 510 Image 520. Specifically, the dental plaque recognition module 152 can convert the ultraviolet light image 510 into the HSV format, and use the fluorescence range threshold (or preset range) to find the position of the dental plaque in the ultraviolet light image 510 . For example, the original ultraviolet image 510 may be in the format of the RGB color space. In addition, the plaque recognition module 152 can convert the ultraviolet image 510 from the RGB color space to the HSV color space. Furthermore, the plaque recognition module 152 can determine a plurality of plaque pixels 521 in the ultraviolet light image 510 after the color space conversion according to a preset range, so as to generate a plaque distribution image 520 as shown in FIG. 5B. In this embodiment, the plaque distribution image 520 in FIG. 5B may be a binary image. Then, the plaque recognition module 152 can stack the plaque distribution image 520 of FIG. 5B and the ultraviolet light image 510 of FIG. 5A to generate a stacked image 530 including a plurality of plaque pixels 531 as shown in FIG. 5C , But the present invention is not limited to this. In one embodiment, the plaque recognition module 152 can stack the plaque distribution image 520 of FIG. 5B and the tooth image 400 (white light image) similar to that of FIG. 4 to generate a white light image of the tooth image. Stacked images of the background, so that it can also be used to show the distribution of dental plaque information. The image analysis system 100 can also display the stacked images 530 through the display device 140 to provide information on the distribution of dental plaque for the medical staff to make judgments. In addition, in other embodiments, the plaque recognition module 152 can convert the ultraviolet light image 510 from the format of the first color space to the format of the second color space, wherein the first color space and the second color space The space can be selected from various color spaces such as RGB color space, HSV color space or HSL color space, and is not limited to the above-mentioned embodiments.

需要說明的是,參考圖3A至圖4,在步驟S211中,附帶一提,在步驟S211中,牙位辨識模組151可選擇性地對白光影像進行影像前處理,以產生牙齒影像400。舉例而言,牙位辨識模組151可對白光影像裁切取下白光影像中牙齦特定區域內包含的部分。並且,牙位辨識模組151可對白光影像進行影像對比調整、亮度調整以及影像正規化處理。類似地,在步驟S221中,牙菌斑辨識模組152可選擇性地對紫外光影像510進行影像前處理,以產生牙菌斑分布影像520。It should be noted that, referring to FIGS. 3A to 4, in step S211, in addition, in step S211, the tooth position recognition module 151 can selectively perform image pre-processing on the white light image to generate the tooth image 400. For example, the tooth position recognition module 151 can crop the white light image to remove the part included in the specific area of the gum in the white light image. In addition, the tooth position recognition module 151 can perform image contrast adjustment, brightness adjustment, and image normalization processing on the white light image. Similarly, in step S221, the plaque recognition module 152 can selectively perform image pre-processing on the ultraviolet image 510 to generate a plaque distribution image 520.

回到圖2,在步驟S230中,處理電路130經由牙位辨識模組151比對牙菌斑分布影像520以及牙齒影像400,以分析多個牙齒區域401-1~401-N中個別的牙菌斑分布。舉例而言,牙位辨識模組151可比對圖5B的牙菌斑分布影像520以及圖4的牙齒影像400,以各別計算所述多個牙齒區域401-1~401-N各別的牙菌斑像素521的數量。接著,在步驟S240中,處理電路130依據上述資訊,建立對應於牙齒的牙菌斑數值表。舉例而言,牙位辨識模組151可依據所述多個牙位的各別的牙菌斑數值來建立牙菌斑數值表。也就是說,牙位辨識模組151可依據所述多個牙位的各別的牙菌斑像素521的數量來計算出牙菌斑數值,並依據牙菌斑數值來建立如以下表1的牙菌斑數值表。 齒號 11 12 13 14 15 16 17 18 牙菌斑數值                 齒號 21 22 22 24 25 26 27 28 牙菌斑數值                 齒號 31 32 33 34 35 36 37 38 牙菌斑數值                 齒號 41 42 43 44 45 46 47 48 牙菌斑數值                 表1 Returning to FIG. 2, in step S230, the processing circuit 130 compares the plaque distribution image 520 and the tooth image 400 through the tooth position identification module 151 to analyze individual teeth in the plurality of tooth regions 401-1 to 401-N. Plaque distribution. For example, the tooth position recognition module 151 can compare the plaque distribution image 520 of FIG. 5B and the tooth image 400 of FIG. The number of plaque pixels 521. Next, in step S240, the processing circuit 130 creates a table of plaque values corresponding to the teeth based on the above information. For example, the tooth position recognition module 151 can create a plaque value table according to the respective plaque values of the multiple tooth positions. That is, the tooth position identification module 151 can calculate the plaque value according to the number of the respective plaque pixels 521 of the plurality of tooth positions, and establish the value as shown in Table 1 below according to the plaque value Table of dental plaque values. Tooth number 11 12 13 14 15 16 17 18 Plaque value Tooth number twenty one twenty two twenty two twenty four 25 26 27 28 Plaque value Tooth number 31 32 33 34 35 36 37 38 Plaque value Tooth number 41 42 43 44 45 46 47 48 Plaque value Table 1

表1是依照本發明一實施例的牙菌斑數值表。表1中的齒號是依據國際牙科聯盟(Fédération Dentaire Internationale,FDI)提出的牙位表示法。具體來說,每顆牙齒用兩位數字表示。第一位數字可用於表示牙齒所在的象限,其中所述象限可例如包括對應於牙齒的右上、左上、左下及右下的位置。並且,第二位數字可用於表示牙齒的位置,例如從中門齒到第三臼齒分別為1-8。此外,表1中的牙菌斑數值是牙位辨識模組151牙位辨識模各別計算對應於各齒號的牙齒區域401-1~401-N中的牙菌斑像素521的數量而得到的數值。Table 1 is a numerical table of dental plaque according to an embodiment of the present invention. The tooth numbers in Table 1 are based on the tooth position notation proposed by the Fédération Dentaire Internationale (FDI). Specifically, each tooth is represented by two digits. The first digit can be used to indicate the quadrant in which the tooth is located, where the quadrant can, for example, include positions corresponding to the upper right, upper left, lower left, and lower right of the tooth. And, the second digit can be used to indicate the position of the teeth, for example, from the middle incisor to the third molar, 1-8 respectively. In addition, the plaque values in Table 1 are obtained by the tooth position recognition module 151, which respectively calculates the number of plaque pixels 521 in the tooth region 401-1~401-N corresponding to each tooth number. The value of.

在本實施例中,處理電路130可透過顯示裝置來顯示如上表1的牙菌斑數值表以及堆疊影像530的至少其中之一,以有效提供牙菌斑分布資訊予醫療人員進行判斷。順帶一提,在一實施例中,影像分析系統100亦可整合至手機等行動電子裝置,以讓使用者可自行檢查牙齒的牙菌斑分布情形。In this embodiment, the processing circuit 130 can display at least one of the plaque value table and the stacked image 530 as shown in Table 1 through the display device, so as to effectively provide plaque distribution information for medical personnel to make judgments. By the way, in one embodiment, the image analysis system 100 can also be integrated into a mobile electronic device such as a mobile phone, so that the user can check the plaque distribution of the teeth by themselves.

綜上所述,本發明的影像分析系統與影像分析方法可自動判斷齒號以及牙菌斑分布情形,有效地減少檢查牙菌斑所需的時間。如此一來,不僅可增加醫療人員的效率,也可減少受測者所許等待的時間。In summary, the image analysis system and image analysis method of the present invention can automatically determine the number of teeth and the distribution of dental plaque, effectively reducing the time required for checking dental plaque. In this way, not only can the efficiency of medical staff be increased, but also the waiting time allowed by the testees can be reduced.

100:影像分析系統 101:白光光源 102:紫外光光源 110:光源驅動電路 120:取像裝置 130:處理電路 140:顯示裝置 150:儲存裝置 151:牙位辨識模組 152:牙菌斑辨識模組 310:正面影像 320、330:側面影像 340:上顎影像 350:下顎影像 400:牙齒影像 401-1~401-N:牙齒區域 510:紫外光影像 520:牙菌斑分布影像 521、531:牙菌斑像素 530:堆疊影像 S210、S211、S220、S221、S230、S240:步驟 100: image analysis system 101: White light source 102: UV light source 110: Light source drive circuit 120: Capture device 130: processing circuit 140: display device 150: storage device 151: Tooth recognition module 152: Dental plaque recognition module 310: Front image 320, 330: Silhouette 340: Upper jaw image 350: jaw image 400: tooth image 401-1~401-N: tooth area 510: Ultraviolet image 520: Image of plaque distribution 521, 531: plaque pixels 530: Stacked images S210, S211, S220, S221, S230, S240: steps

圖1是依照本發明的一實施例的一種影像分析系統的示意圖。 圖2是依照本發明的一實施例的影像分析方法的流程圖。 圖3A是依照本發明的一實施例的牙齒的頰側的正面影像的示意圖。 圖3B是依照本發明的一實施例的牙齒的頰側的一側面影像的示意圖。 圖3C是依照本發明的一實施例的牙齒的頰側的另一側面影像的示意圖。 圖3D是依照本發明的一實施例的牙齒的咬合側的上顎影像的示意圖。 圖3E是依照本發明的一實施例的牙齒的咬合側的下顎影像的示意圖。 圖4是依照本發明的一實施例的牙位辨識的示意圖。 圖5A是依照本發明的一實施例的紫外光影像的示意圖。 圖5B是依照本發明的一實施例的牙菌斑分布影像的示意圖。 圖5C是依照本發明的一實施例的堆疊影像的示意圖。 FIG. 1 is a schematic diagram of an image analysis system according to an embodiment of the invention. FIG. 2 is a flowchart of an image analysis method according to an embodiment of the invention. 3A is a schematic diagram of a frontal image of the buccal side of a tooth according to an embodiment of the present invention. 3B is a schematic diagram of a profile image of the buccal side of a tooth according to an embodiment of the present invention. FIG. 3C is a schematic diagram of another silhouette image of the buccal side of a tooth according to an embodiment of the present invention. 3D is a schematic diagram of an image of the upper jaw on the occlusal side of a tooth according to an embodiment of the present invention. 3E is a schematic diagram of a lower jaw image of the occlusal side of a tooth according to an embodiment of the present invention. Fig. 4 is a schematic diagram of tooth position identification according to an embodiment of the present invention. FIG. 5A is a schematic diagram of an ultraviolet light image according to an embodiment of the invention. 5B is a schematic diagram of an image of dental plaque distribution according to an embodiment of the present invention. FIG. 5C is a schematic diagram of stacked images according to an embodiment of the invention.

100:影像分析系統 100: image analysis system

101:白光光源 101: White light source

102:紫外光光源 102: UV light source

110:光源驅動電路 110: Light source drive circuit

120:取像裝置 120: Capture device

130:處理電路 130: processing circuit

140:顯示裝置 140: display device

150:儲存裝置 150: storage device

151:牙位辨識模組 151: Tooth recognition module

152:牙菌斑辨識模組 152: Dental plaque recognition module

Claims (18)

一種影像分析系統,包括:一儲存裝置,用以儲存一牙位辨識模組以及一牙菌斑辨識模組;一取像裝置,用以取得對應於一牙齒的一白光影像以及一紫外光影像;以及一處理電路,耦接該儲存裝置以及該取像裝置,其中該處理電路將該白光影像輸入至該牙位辨識模組,並且該牙位辨識模組輸出被定義有對應於該牙齒的多個牙位的多個牙齒區域的一牙齒影像,其中該牙位辨識模組對該白光影像進行一影像前處理,以產生該牙齒影像,並且該牙位辨識模組依據一牙齒資料庫對該牙齒影像進行牙位標記,以藉由影像辨識該牙齒影像中的多個牙齒特徵來定義在該牙齒影像中的該些牙位的該些牙齒區域,其中該處理電路將該紫外光影像輸入至該牙菌斑辨識模組,並且該牙菌斑辨識模組依據該紫外光影像來輸出一牙菌斑分布影像,其中該處理電路藉由該牙位辨識模組比對該牙菌斑分布影像以及該牙齒影像,以分析該些牙齒區域中各別的一牙菌斑分布,並且建立對應於該牙齒的一牙菌斑數值表。 An image analysis system includes: a storage device for storing a tooth position recognition module and a dental plaque recognition module; an image capturing device for obtaining a white light image and an ultraviolet light image corresponding to a tooth And a processing circuit coupled to the storage device and the image capturing device, wherein the processing circuit inputs the white light image to the tooth position identification module, and the tooth position identification module output is defined to correspond to the tooth A tooth image of a plurality of tooth regions of a plurality of tooth positions, wherein the tooth position recognition module performs an image pre-processing on the white light image to generate the tooth image, and the tooth position recognition module compares the teeth according to a tooth database The tooth image is marked with a tooth position to define the tooth regions of the tooth positions in the tooth image by identifying a plurality of tooth features in the tooth image by the image, wherein the processing circuit inputs the ultraviolet light image To the dental plaque recognition module, and the dental plaque recognition module outputs a dental plaque distribution image according to the ultraviolet light image, wherein the processing circuit compares the dental plaque distribution with the dental position recognition module The image and the tooth image are used to analyze the distribution of a plaque in each of the tooth regions, and establish a plaque value table corresponding to the tooth. 如請求項1所述的影像分析系統,更包括:一白光光源,耦接該處理電路;以及 一紫外光光源,耦接該處理電路,其中當該處理電路操作該白光光源照明該牙齒時,該處理電路透過該取像裝置拍攝該牙齒,以取得該白光影像,其中當該處理電路操作該紫外光光源照明該牙齒時,該處理電路透過該取像裝置拍攝該牙齒,以取得該紫外光影像,其中該取像裝置在不同時間從對應於該牙齒的同一取樣位置來取得該白光影像以及該紫外光影像。 The image analysis system according to claim 1, further comprising: a white light source coupled to the processing circuit; and An ultraviolet light source is coupled to the processing circuit, wherein when the processing circuit operates the white light source to illuminate the tooth, the processing circuit captures the tooth through the imaging device to obtain the white light image, wherein when the processing circuit operates the tooth When the ultraviolet light source illuminates the tooth, the processing circuit captures the tooth through the imaging device to obtain the ultraviolet light image, wherein the imaging device obtains the white light image from the same sampling position corresponding to the tooth at different times, and The UV image. 如請求項2所述的影像分析系統,其中該紫外光光源的亮度值大於該白光光源的亮度值。 The image analysis system according to claim 2, wherein the brightness value of the ultraviolet light source is greater than the brightness value of the white light source. 如請求項1所述的影像分析系統,其中該取像裝置用以從對應於該牙齒的多個不同取樣位置來取得多組白光影像以及紫外光影像。 The image analysis system according to claim 1, wherein the image capturing device is used to obtain multiple sets of white light images and ultraviolet light images from a plurality of different sampling positions corresponding to the tooth. 如請求項1所述的影像分析系統,其中該牙菌斑辨識模組將該紫外光影像從一RGB色彩空間轉換至一HSV色彩空間,並且依據一預設範圍來判斷經色彩空間轉換後的該紫外光影像中的多個牙菌斑像素,以產生該牙菌斑分布影像。 The image analysis system according to claim 1, wherein the plaque recognition module converts the ultraviolet light image from an RGB color space to an HSV color space, and determines the converted color space according to a preset range A plurality of plaque pixels in the ultraviolet light image are used to generate the plaque distribution image. 如請求項5所述的影像分析系統,其中該牙菌斑分布影像為一二值化影像,並且該牙位辨識模組比對該牙菌斑分布影像以及該牙齒影像,以各別計算該些牙齒區域各別的一牙菌斑像素數量,其中該牙位辨識模組依據該些牙位的各別的該牙菌斑數值來建立該牙菌斑數值表。 The image analysis system according to claim 5, wherein the plaque distribution image is a binary image, and the tooth position recognition module compares the plaque distribution image with the tooth image to calculate the The number of plaque pixels in each of the tooth regions, wherein the tooth position identification module establishes the plaque value table according to the respective plaque values of the tooth positions. 如請求項1所述的影像分析系統,其中該影像分析系統透過一顯示裝置顯示該牙菌斑數值表。 The image analysis system according to claim 1, wherein the image analysis system displays the dental plaque value table through a display device. 如請求項1所述的影像分析系統,其中該牙菌斑辨識模組將該牙菌斑分布影像以及該紫外光影像進行影像堆疊,以產生一堆疊影像,並且該影像分析系統透過一顯示裝置顯示該堆疊影像。 The image analysis system according to claim 1, wherein the plaque recognition module performs image stacking of the plaque distribution image and the ultraviolet image to generate a stacked image, and the image analysis system transmits through a display device Display the stacked image. 如請求項1所述的影像分析系統,其中該取像裝置取得該紫外光影像的曝光時間長度大於取得該白光影像的曝光時間長度。 The image analysis system according to claim 1, wherein the exposure time length for the ultraviolet light image obtained by the image capturing device is longer than the exposure time length for the white light image obtained by the image capturing device. 一種影像分析方法,包括:透過一取像裝置取得對應於一牙齒的一白光影像以及一紫外光影像;將該白光影像輸入至一牙位辨識模組,以使該牙位辨識模組輸出被定義有對應於該牙齒的多個牙位的多個牙齒區域的一牙齒影像,包括:對該白光影像進行一影像前處理,以產生該牙齒影像;以及依據一牙齒資料庫對牙齒影像進行牙位標記,並且影像辨識該牙齒影像中的多個牙齒特徵來定義在該牙齒影像中的該些牙位的該些牙齒區域;將該紫外光影像輸入至一牙菌斑辨識模組,以使該牙菌斑辨識模組依據該紫外光影像來輸出一牙菌斑分布影像;以及 藉由該牙位辨識模組比對該牙菌斑分布影像以及該牙齒影像,以分析該些牙齒區域中各別的一牙菌斑分布,並且建立對應於該牙齒的一牙菌斑數值表。 An image analysis method includes: obtaining a white light image and an ultraviolet light image corresponding to a tooth through an image capturing device; Defining a tooth image of a plurality of tooth regions corresponding to a plurality of tooth positions of the tooth includes: performing an image pre-processing on the white light image to generate the tooth image; and performing a tooth image on the tooth image according to a tooth database Position mark, and the image recognizes multiple tooth features in the tooth image to define the tooth regions of the teeth in the tooth image; input the ultraviolet light image to a plaque recognition module to make The dental plaque recognition module outputs a dental plaque distribution image according to the ultraviolet light image; and The plaque distribution image and the tooth image are compared by the tooth position recognition module to analyze the plaque distribution in each of the tooth regions, and create a plaque value table corresponding to the tooth . 如請求項10所述的影像分析方法,其中取得對應於該牙齒的該白光影像以及該紫外光影像的步驟包括:當一白光光源照明該牙齒時,透過該取像裝置拍攝該牙齒,以取得該白光影像;以及當一紫外光光源照明該牙齒時,透過該取像裝置拍攝該牙齒,以取得該紫外光影像,其中該取像裝置在不同時間從對應於該牙齒的同一取樣位置來取得該白光影像以及該紫外光影像。 The image analysis method according to claim 10, wherein the step of obtaining the white light image and the ultraviolet light image corresponding to the tooth comprises: when a white light source illuminates the tooth, the tooth is photographed through the image capturing device to obtain The white light image; and when an ultraviolet light source illuminates the tooth, the tooth is photographed through the imaging device to obtain the ultraviolet image, wherein the imaging device obtains the ultraviolet image from the same sampling position corresponding to the tooth at different times The white light image and the ultraviolet light image. 如請求項11所述的影像分析方法,其中該紫外光光源的亮度值大於該白光光源的亮度值。 The image analysis method according to claim 11, wherein the brightness value of the ultraviolet light source is greater than the brightness value of the white light source. 如請求項10所述的影像分析方法,其中該取像裝置用以從對應於該牙齒的多個不同取樣位置來取得多組白光影像以及紫外光影像。 The image analysis method according to claim 10, wherein the image capturing device is used to obtain multiple sets of white light images and ultraviolet light images from a plurality of different sampling positions corresponding to the tooth. 如請求項10所述的影像分析方法,其中將該紫外光影像輸入至該牙菌斑辨識模組,以使該牙菌斑辨識模組依據該紫外光影像來輸出該牙菌斑分布影像的步驟包括:將該紫外光影像從一RGB色彩空間轉換至一HSV色彩空間;以及依據一預設範圍來判斷經色彩空間轉換後的該紫外光影像中 的多個牙菌斑像素,以產生該牙菌斑分布影像。 The image analysis method according to claim 10, wherein the ultraviolet light image is input to the dental plaque recognition module, so that the dental plaque recognition module outputs the distribution image of the dental plaque according to the ultraviolet light image The steps include: converting the ultraviolet light image from an RGB color space to an HSV color space; and judging the ultraviolet light image after the color space conversion according to a preset range A plurality of plaque pixels of the plaque to produce an image of the plaque distribution. 如請求項14所述的影像分析方法,其中該牙菌斑分布影像為一二值化影像,並且建立對應於該牙齒的該牙菌斑數值表的步驟包括:比對該牙菌斑分布影像以及該牙齒影像,以各別計算該些牙齒區域各別的一牙菌斑像素數量;以及依據該些牙位的各別的該牙菌斑數值來建立該牙菌斑數值表。 The image analysis method according to claim 14, wherein the plaque distribution image is a binary image, and the step of establishing the plaque value table corresponding to the tooth includes: comparing the plaque distribution image And the tooth image is used to separately calculate the number of plaque pixels in the tooth regions; and the plaque value table is established according to the respective plaque values of the tooth positions. 如請求項10所述的影像分析方法,更包括:透過一顯示裝置顯示該牙菌斑數值表。 The image analysis method according to claim 10, further comprising: displaying the dental plaque value table through a display device. 如請求項10所述的影像分析方法,更包括:藉由該牙菌斑辨識模組將該牙菌斑分布影像以及該紫外光影像進行影像堆疊,以產生一堆疊影像;以及透過一顯示裝置顯示該堆疊影像。 The image analysis method according to claim 10, further comprising: image stacking the plaque distribution image and the ultraviolet light image by the plaque recognition module to generate a stacked image; and through a display device Display the stacked image. 如請求項10所述的影像分析方法,其中該取像裝置取得該紫外光影像的曝光時間長度大於取得該白光影像的曝光時間長度。 The image analysis method according to claim 10, wherein the exposure time length for the ultraviolet light image obtained by the image capturing device is longer than the exposure time length for the white light image obtained by the image capturing device.
TW109141592A 2020-11-26 2020-11-26 Image analysis system and image analysis method TWI749909B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW109141592A TWI749909B (en) 2020-11-26 2020-11-26 Image analysis system and image analysis method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW109141592A TWI749909B (en) 2020-11-26 2020-11-26 Image analysis system and image analysis method

Publications (2)

Publication Number Publication Date
TWI749909B true TWI749909B (en) 2021-12-11
TW202221647A TW202221647A (en) 2022-06-01

Family

ID=80681281

Family Applications (1)

Application Number Title Priority Date Filing Date
TW109141592A TWI749909B (en) 2020-11-26 2020-11-26 Image analysis system and image analysis method

Country Status (1)

Country Link
TW (1) TWI749909B (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080058786A1 (en) * 2006-04-12 2008-03-06 Searete Llc, A Limited Liability Corporation Of The State Of Delaware Autofluorescent imaging and target ablation
TWM503883U (en) * 2015-03-30 2015-07-01 Dentmate Technology Co Ltd Dental plaque detector with camera
CN105942962A (en) * 2016-06-05 2016-09-21 郭菲 Oral cavity endoscope diagnosis supporting device
TW202009865A (en) * 2018-08-30 2020-03-01 廣達電腦股份有限公司 Image processing method and device for fluorescence region of teeth
TW202014151A (en) * 2018-10-10 2020-04-16 歐陽聰星 Method, system, computer device and storage medium of virtual editing of teeth

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080058786A1 (en) * 2006-04-12 2008-03-06 Searete Llc, A Limited Liability Corporation Of The State Of Delaware Autofluorescent imaging and target ablation
TWM503883U (en) * 2015-03-30 2015-07-01 Dentmate Technology Co Ltd Dental plaque detector with camera
CN105942962A (en) * 2016-06-05 2016-09-21 郭菲 Oral cavity endoscope diagnosis supporting device
TW202009865A (en) * 2018-08-30 2020-03-01 廣達電腦股份有限公司 Image processing method and device for fluorescence region of teeth
TW202014151A (en) * 2018-10-10 2020-04-16 歐陽聰星 Method, system, computer device and storage medium of virtual editing of teeth

Also Published As

Publication number Publication date
TW202221647A (en) 2022-06-01

Similar Documents

Publication Publication Date Title
US11810271B2 (en) Domain specific image quality assessment
US11452577B2 (en) Generation of synthetic post treatment images of teeth
US7454046B2 (en) Method and system for analyzing skin conditions using digital images
CN104470416B (en) Image processing apparatus and endoscope apparatus
CN107800966B (en) Method, apparatus, computer readable storage medium and the electronic equipment of image procossing
KR102176490B1 (en) Methods for Segmentation and Processing of Dental Images for Diagnosing, Predicting, or Managing Oral Conditions
US20080232662A1 (en) Outline detection apparatus, outline detection method, and program thereof
CN113379697B (en) Color image caries identification method based on deep learning
ALbahbah et al. Detection of caries in panoramic dental X-ray images using back-propagation neural network
Koutsouri et al. Detection of occlusal caries based on digital image processing
TWI749909B (en) Image analysis system and image analysis method
Chen et al. Automated Detection System Based on Convolution Neural Networks for Retained Root, Endodontic Treated Teeth, and Implant Recognition on Dental Panoramic Images
CN114399621A (en) Oral disease recognition system based on deep learning
Liu et al. Automatic detection and classification of dental fluorosis in vivo using white light and fluorescence imaging
Ghaznavi Bidgoli et al. Automatic diagnosis of dental diseases using convolutional neural network and panoramic radiographic images
Javid et al. Marking early lesions in labial colored dental images using a transfer learning approach
CN115552512A (en) Image display system and image display method
IL264041A (en) Personal authentication device
KR102636215B1 (en) Electronic device, method and program for scanning scalp based on image
CN111462114A (en) Tooth color value determination method and device and electronic equipment
JP2008244794A (en) Image processor, and image processing method
Keerthana et al. Classification of tooth type from dental X-ray image using projection profile analysis
JP2016151584A (en) Organ image capturing device
KR102434843B1 (en) Artificial teeth manufacturing information generation method and artificial teeth manufacturing system
WO2022095839A1 (en) Image processing method, and electronic device and readable storage medium