CN1315102C - Volume-transparent roaming method and system for objects - Google Patents

Abstract

The invention relates to a volume transparent roaming method of an object and its system, mainly based on the development of the visualization technology of volume data to realize the transparent visualization level of the three-dimensional volume structure, combined with the interactive virtual roaming technology of the object to form a unified technologies and systems. The technical feature lies in: performing transparent processing on three-dimensional object data to form transparent data on three-dimensional objects, performing volume rendering processing on the above-mentioned transparent volume data, obtaining a transparent three-dimensional virtual body structure, and using virtual roaming technology for interaction Through roaming processing, it realizes interactive volume-transparent roaming from the whole interior to any direction outside or from the outside to any direction inside. Utilizing the present invention, after endowing the inspector with the "penetrating" ability of vision, the inspector can virtually perform roaming observation and detection capabilities and three-dimensional precise surgical navigation.

Description

Volume-transparent roaming method and system for objects

Technical field: The present invention relates to a volume transparent roaming method of an object and its system, which is mainly based on the development and realization of the volume data (VolumeData) visualization technology to the level of three-dimensional volume transparency, combined with the interactive virtual roaming technology of the object And become a unified technology and system—all-round, interactive body-transparent roaming observation, detection technology and system for object internal peeping (viewing) and external viewing.

Background technology: The "transparency" mentioned in the present invention strictly refers to the intermediate state between "full opaque" and "full transparency", or "semi-transparent". Because, if the observed object is really fully transparent, nothing can be seen, and there is no practical significance. According to the literature search, there is no report on the transparent observation and detection of the technologies and systems for virtual roaming observation and detection of three-dimensional objects or their parts. For example, the literature "Key Technologies in Interactive Real-time Virtual Endoscopy System" (-see "Computer Application" Volume 22, No. 11, 2002) and "Remote Interactive Virtual Endoscopy System" (-see "Computer Application" Engineering" Vol. 28, No. 11, 2002) in the medical virtual endoscope, although the observer can perform virtual roaming inside the cavity of the human body (such as the stomach), but because of its lack of body transparency, so It can only observe the shape and condition of the epidermis of the cavity, but cannot see the thickness and structure of the cavity wall, let alone observe the three-dimensional structure of the human body from the inside of the cavity to the outside (back) of the cavity wall to the outer surface of the human body. Or the three-dimensional reality of some kind of organization.

In addition, the data transparency processing in existing technologies such as "Development and Theoretical Research of Three-dimensional Human Virtual Endoscopy" (-see "Life Science Research" Vol. 6, No. 4, 2002) is simple and low. The method of directly assigning a certain transparency value to data according to its gray value has not been connected with human visual habit factors such as viewpoint distance. Therefore, the follow-up volume transparency effect cannot meet the actual use requirements, especially the transparency here is not connected or combined with the virtual roaming technology conceptually and technically, so this transparent observation method can only be a fixed viewpoint , Observation in a fixed direction, it is impossible to roam in any direction. It is also impossible to highlight certain details, dilute, peel off or ignore certain details and backgrounds to achieve a transparent roaming observation of a certain organization in a three-dimensional volume.

Due to the respective defects and limitations of the above technologies, it is neither related nor difficult to achieve a full range, interactive virtual roaming, visual "Penetrating" three-dimensional transparent observation and detection.

Since the position and detection ability of the detector cannot be carried out in any direction along any path inside the object, or from the outside of the object to any direction inside the object, the detection accuracy and detection efficiency cannot be effectively improved, nor can it It is convenient for multi-person consultation or remote consultation.

Summary of the invention: In order to overcome and make up for the defects and limitations of the prior art, the present invention proposes a method and system for volume-transparent roaming of objects. Convert the surface data of a series of slices of 3D objects obtained by CT, MRI and other detectors into volume data, and perform volume data transparency processing and 3D reconstruction based on volume rendering to obtain 3D transparent virtual objects corresponding to the original 3D objects On the basis of the transparent objects, the virtual roaming process is carried out, and finally the volume transparent interactive roaming technology and method for the objects are formed.

Utilizing the present invention, the "penetration" ability of the inspector's vision is endowed, or it is equivalent to extending and expanding the original three-dimensional object's surface three-dimensional perception ability of human vision to the depth field of view to the insight into the internal structure of the three-dimensional object At the same time as the perspective ability, it also gives the inspector the behavior ability to virtually walk along any path to any direction inside the three-dimensional object, or perform roaming observation and inspection from the outside to the inside of the three-dimensional object.

The application and embodiment of the present invention in the field of medicine and its clinical practice is to make the computer-aided diagnosis and navigation based on the doctor's visual observation, from the diagnosis based on the image and the navigation based on the plane image, in terms of performance, mode and level, extend and The leap to a new type of three-dimensional transparent roaming auxiliary diagnosis and precise navigation is to a certain extent a promotion of modern medical civilization.

Technical feature of the present invention is:

Carry out volume-based volume transparency processing on the volume data of three-dimensional objects to form volume transparency data of the object; then use volume rendering technology to perform volume rendering processing on the above-mentioned transparent volume data, that is, 3D reconstruction based on volume rendering, Obtain a three-dimensional virtual object corresponding to the original real object with a transparent internal structure; then use the virtual roaming technology to perform interactive roaming processing on it, so as to realize any orientation along any path inside the object or its part, or by Interactive volume-transparent roaming from outside to inside in any direction.

For the plane data of a series of slices of three-dimensional real objects obtained by CT, MRI or other detectors or the data in DICOM format, preprocessing-data conversion is performed first. The data conversion method may be to directly superimpose the surface data vertically, so that the two-dimensional array becomes a three-dimensional array, thereby converting the surface data into volume data. On this basis, this kind of volume data can also be supplemented with data generalized fuzzy enhancement processing to form high-quality volume data with enhanced gray contrast.

The volume data of the (three-dimensional) object also includes individual volume data of various parts (structures) of the object: that is, in the generated volume data (set) of the object, according to the data characteristics of each local (structure), The volume data conforming to these characteristics are extracted to form the volume data (subset) of each part (structure).

A system for realizing the above-mentioned transparent roaming method for objects, characterized in that: the system includes a volume data transparent processor 4, a three-dimensional transparent virtual object generator 5, and a virtual roaming processor 7;

Volume data transparency processor 4: perform volume transparency processing on (3D) volume data of 3D objects or some interesting local volume data to form volume data with volume transparency; the system adopts volume transparency based on volume Treatment techniques and scenarios. The specific implementation method can adopt the direct assignment method of voxel transparency value, or the multi-factor fusion assignment method of voxel transparency value;

3D transparent virtual object generator 5: For the transparentized volume data of the object or the partial volume of the object, use the volume rendering technology to perform volume rendering processing, that is, the 3D reconstruction work based on volume rendering, to obtain the corresponding to the original real object, A three-dimensional virtual object or part thereof whose internal structure is transparent;

Virtual roaming processor 7: For the above-mentioned transparent virtual three-dimensional object or its parts, use the virtual roaming technology to perform interactive roaming processing, so as to realize any direction along any path inside the object or its part or some organization, or from the outside Interactive volumetric transparent roaming in any direction inside. This provides a new type of volume-transparent roaming method and mechanism for object detection, diagnosis or medical navigation based on visual observation.

The volume transparency processing technology and scheme based on volume can adopt the current voxel transparency (value) simple value assignment method, or the voxel transparency (value) multi-factor fusion value method developed by the present invention (this project). Among them, the voxel transparency (value) simple assignment method is the current processing method, that is, a transparency value T=K ₁ T ₁ is assigned in 256 levels of transparency to each voxel in the volume data.

The multi-factor fusion and assignment method of volume (pixel) transparency (value) is: set the transparency value T=K ₁ T ₁ +K ₂ T ₂ +K ₃ T ₃ , wherein: K ₁ , K ₂ , K ₃ can be The weighting factor is an integer or percentage within a certain value range; T ₁ : the original gray value of a voxel, T ₂ : the Euclidean distance between the viewpoint and the voxel, T ₃ : the light source and the voxel Euclidean distance of . Since this method takes T ₂ , T ₃ and other factors related to people's visual habits into consideration, the voxels processed by the volume data transparency processor of this method have detailed and comprehensive transparency assignments, so that the overall transparency effect is clearer and more real The feeling is stronger, and the proportion of each factor is flexible and adjustable.

Before the transparent processor of the volume data, the surface data of a series of slices obtained by CT, MRI or other detectors or the data in DICOM format must be converted into volume data, which requires the addition of a data converter 2 . The data converter 2 is: directly stacking the area data vertically so that the data changes from a two-dimensional array to a three-dimensional array, that is, converting the area data into volume data. On this basis, this kind of volume data can also be supplemented with data generalized fuzzy enhancement processing. That is to say, either the direct method or the direct method plus data generalized fuzzy enhancement can be used.

The direct method is the usual and simplest volume data formation method, because it directly superimposes the area data vertically so that the data changes from a two-dimensional array to a three-dimensional array, so it is called the direct conversion method from area data to volume data. referred to as the direct method.

A local volume data extractor 3 (for three-dimensional objects) and a roaming detection site and roaming mode selector 8 can be added to the system. The local volume data extractor 3 of the (three-dimensional object) is: according to the data characteristics of each local structure or tissue of interest, the relevant volume data is extracted to form the volume data of each local or tissue The roaming detection site and roaming mode selector 8 are: controllable continuous or controllable step-by-step observation mode can be selected; Zoom refinement and enlargement observation mode or non-Zoom normal observation mode can be selected. Including volume transparent controllable continuous roaming 9, volume transparent controllable stepping roaming 10 and transparent Zoom controllable continuous roaming 11 of a part of an object or a certain organization.

In order to facilitate the users of this system to choose different transparency under different objects, different situations, and different habits, so as to improve the accuracy and speed of detection (or navigation), after the three-dimensional transparent virtual object generator 5, volume transparency contrast enhancement can be added Device 6: It is the same principle as in data converter 2 to perform data generalized fuzzy enhancement processing on data with only gray values to enhance its gray contrast. Here, data generalized fuzzy enhancement processing on volume data with transparency values can enhance Its transparent contrast. Use this volume transparency enhancement technology to enhance the transparency contrast, so as to achieve the purpose of highlighting certain details, or diluting, stripping, ignoring certain details and backgrounds, and improving the clarity of volume transparency roaming.

Description of drawings:

Figure 1: Realization and Composition Block Diagram of Volume Transparent Roaming System

Figure 2: Transparency rendering of the human brain body

a: The original appearance of the non-transparent human brain

b: low-transparency transparent human brain map

c: High-transparency transparent human brain map

Figure 3: Transparency rendering of the skull body

a: The original appearance of the non-transparent skull

b: low-transparency body transparent head map

c: Transparency body transparent head map

d: High-transparency body transparent head map

Detailed ways:

The present invention will be further described now in conjunction with accompanying drawing:

Utilizing the present invention, on the basis of making the object to be examined and diagnosed or the human body (or human tissue) undergoing surgical treatment to be in a three-dimensional structural transparent state, the system user can virtually go to any direction along any path inside the object, or from Perform volume-transparent roaming observation, diagnosis or medical navigation from the outside of the object to any direction inside.

In this embodiment, based on MRI and various CT tomographic slice data or data in DICOM format, the non-transparent human brain map 2(a) and the non-transparent skull map 3(a) are taken as examples. Combining with Figure 1, the implementation and composition of the volumetric transparent roaming system (device) are dissected.

CT, MRI and other detection data or DICOM format data 1 pass through the data converter 2 to convert a series of slice (two-dimensional) surface data of the object to form volume data of the object. The data conversion method can be to directly superimpose the area data vertically so that the data changes from a two-dimensional array to a three-dimensional array, that is, the area data is converted into volume data. On this basis, this kind of volume data can also be supplemented with data generalized fuzzy enhancement processing.

Utilize the local volume data extractor 3 of the three-dimensional object, according to the data feature of each local structure, extract the volume data related to it, and form the (three-dimensional) volume data of each local or organization;

In the volume data transparent processor 4, adopt volume (pixel) transparency (value) multi-factor fusion assignment method to carry out volume transparency processing to the volume data of object or its each partial volume data, form volume transparency data;

Using volume rendering technology for volume rendering processing, that is, 3D reconstruction based on volume rendering, and obtaining a transparent 3D virtual object corresponding to the real object through the 3D transparent virtual object generator 5;

The Volume Transparency Contrast Enhancer 6 employs data generalized blur enhancement processing to enhance transparency contrast.

For the above-mentioned transparent virtual three-dimensional object or part thereof, the virtual roaming processor 7 is used to perform interactive roaming processing, so as to realize moving from the object or its part or a certain organization along any path to any direction, or from the outside to any direction inside. Interactive volumetric transparent walkthrough. This provides a new way and mechanism of volume-transparent roaming for visual detection, diagnosis or medical navigation of objects.

Claims (3)

1. the transparent loaming method of the body of an object, it is characterized in that: the volume data to three-dimensional body is carried out the body transparent processing with the multifactor fusion assignment of voxel transparency method, the multifactor fusion assignment of described voxel transparency method: make transparence value T=K ₁T ₁+ K ₂T ₂+ K ₃T ₃, wherein: K ₁, K ₂, K ₃Be adjustable factors; T ₁: be gray-scale value, the T of voxel ₂: the Euclidean distance of viewpoint and voxel, T ₃: the Euclidean distance of light source and voxel; Through the processing of the multifactor fusion assignment of described voxel transparency method, form transparent volume data; To above-mentioned transparent volume data, utilize Volume Rendering Techniques to carry out volume drawing, that is based on the three-dimensional reconstruction work of volume drawing, obtain three-dimensional transparent dummy object: to above-mentioned three-dimensional transparent dummy object with the corresponding inner structure transparence of former real-world object, utilizing the virtual roaming technology to carry out the transparent roaming of continuous interactive formula handles, realization from the inside of object along free routing to arbitrary orientation, or by the interactive body transparent roaming of outside to inner arbitrary orientation.

2. system that realizes the transparent loaming method of body of above-mentioned object, it is characterized in that: this system comprises that volume data transparent processing device (4), three-dimensional transparent substance form device (5), virtual roaming processor (7);

Volume data transparent processing device (4): volume data or its local volume data to three-dimensional body are carried out the body transparent processing, form transparent volume data; System adopts body transparent processing technology and the scheme based on body, and the specific implementation method adopts the multifactor fusion assignment of voxel transparency method:

The three-dimensional transparent substance forms device (5): to object or its local transparent volume data, utilize Volume Rendering Techniques to carry out volume drawing, that is, obtain three-dimensional transparent dummy object or its partial structurtes with the corresponding inner structure transparence of former real-world object based on the three-dimensional reconstruction work of volume drawing;

Virtual roaming processor (7): to above-mentioned three-dimensional transparent dummy object or its partial structurtes, utilizing the virtual roaming technology to carry out the roaming of continuous interactive formula handles, realization from the inside of object along free routing to arbitrary orientation, or by the interactive body transparent roaming of outside to inner arbitrary orientation.

3. system according to claim 2, it is characterized in that: add roaming detection position and roaming mode selector (8) in system, described roaming detects the position and roams the transparent Zoom controllable continuous roaming (11) that mode selector (8) comprises the transparent controllable continuous formula roaming of body (9), the transparent controlled step-by-step movement roaming of body (10) and object partial structurtes.

Images