TW201427643A - Image capturing apparatus and capturing method - Google Patents

Abstract

An image capturing apparatus including a plurality of image sensing modules and at least a light source is provided for image capturing an eye. Each of the image sensing modules includes an image sensor and a lens. The light source emits an illumination light, and the illumination light irradiates the eye. The eye reflects the illumination light into an image light. The image light includes a plurality of image sub-beams, and the image sub-beams are transmitted to the image sensors of the image sensing modules through the lenses of the image sensing modules, respectively. A detecting method is also provided.

Description

Image capturing device and reading method

The present invention relates to an image capturing device and a capturing method, and more particularly to an image capturing device for capturing an eyeball and a capturing method.

The eyes are the window of the soul, through which people can feel the light and color of the world. The cone-shaped cells and rod-shaped cells that sense color and light in the eye are located on the retina of the fundus, and are the only tissues in the human body that can convert light into physiological electrical signals. Among them, the blood vessels that supply blood and nutrients to the eyes are also located at the fundus. When there is vascular hyperplasia or rupture in the fundus, such as symptoms such as macular hemorrhage, it is easy to cause cone and cell death on the retina, which causes the patient to lose vision. Therefore, in the diagnosis and prevention of eye diseases, the observation and tracking of the fundus images is extremely important.

In general, due to the limitation of the pupil size of the human eye, the traditional fundus imaging method can only capture a range of fundus images in a single shot at one angle, even if a pupil dilating drug such as a mydriatic agent is used. It is a 30 to 40 degree angle of view. Therefore, if you want to take an image at the edge of the fundus, the patient will always stare at an observation reference point, and then move the point where the eye is gazing at a slow, steady and sustained speed, up and down, and so on. The image of the fundus. Then, using a data processing device such as a computer, a plurality of images of the fundus are combined through a subsequent dedicated image synthesizing software. However, the illumination light when shooting multiple fundus images repeatedly will also The patient's eyes feel uncomfortable and tired, which may cause unconscious blinking or eyeball tremors that affect the quality of the camera. In addition, since the images of these fundus are obtained by multiple shots, the exposure value and white balance of each shot are not the same, and these images need to be corrected first, so that a large computer such as a computer can be used for splicing. The difficulty of the correction increases and the image quality after stitching is affected. If the quality of the stitched image is affected, it is easy for the medical staff to identify the microvascular image of the fundus, which may cause difficulty in judgment or even delay the treatment of the patient. Therefore, how to quickly obtain a more complete and clear fundus image is a current priority for medical care.

The invention provides an image capturing device capable of capturing images of a plurality of different regions of an eyeball.

The present invention provides a method of capturing, which can simultaneously capture multiple images of an eyeball from a plurality of different angles and combine the images.

The invention provides an image capturing device for capturing an image of an eyeball, the image detecting device comprising a plurality of image sensing modules and at least one light source. Each image sensing module includes an image sensor and a lens. The light source emits an illumination light, and the illumination light illuminates the eyeball, and the eyeball reflects the illumination light into an image light. The image light includes a plurality of sub-image light beams, and the sub-image light beams are respectively transmitted to the image senses through the lenses of the image sensing modules. Detector.

In an embodiment of the invention, the illumination light is irradiated to the fundus of the eyeball through the pupil of the eyeball, and the fundus reflects the illumination light into the image light, and the shadow These sub-image beams of light are transmitted to the image sensing modules via the pupils, respectively.

In an embodiment of the present invention, the image capturing range of the fundus is partially overlapped by the adjacent two image sensing modules.

In an embodiment of the invention, the optical axes of the lenses of the image sensing modules are not parallel to each other, and the optical axes of the lenses pass through the pupils of the eyeballs.

In one embodiment of the present invention, each of the image sensing modules further includes an actuator coupled to at least one of the image sensor and the lens to focus the image sensing module.

In an embodiment of the present invention, each of the image sensing modules further includes a micro processing unit electrically connected to the corresponding image sensor to extract the sub image beam measured by the image sensor. The data of the resulting image.

In an embodiment of the present invention, the image capturing device further includes a processing unit electrically connected to the image sensing modules to respectively generate the sub-image beams measured by the image sensors. The images of multiple eyeballs merge.

In an embodiment of the invention, the images of the eyeballs measured by the adjacent two image sensors are partially overlapped.

In one embodiment of the invention, the processing unit compares the overlapping portions of the images as a correction reference when combining the images.

In one embodiment of the present invention, the processing unit includes a first comparison module, a second comparison module, a third comparison module, a fourth comparison module, and a determination module. The first comparison module takes these peripheral images One of the first peripheral images is compared with the overlap of the central image. The second comparison module compares the overlap between one of the peripheral images and the central image. The third comparison module calculates an average image of the overlapping portion of the first peripheral image and the second peripheral image, and compares the average image with the overlapping portion of the central image. The fourth comparison module calculates a gradation image of the overlapping portion of the first peripheral image and the second peripheral image, and compares the overlapping portion of the gradation image with the central image. The judging module judges whether the comparison result of the first to fourth comparison modules is the smallest, wherein when the comparison difference of the first comparison module is the smallest, the judging module adopts the first part of the overlap between the first peripheral image and the second peripheral image. Data of a peripheral image; when the comparison difference of the second comparison module is the smallest, the judging module uses the data of the second peripheral image for the overlapping portion of the first peripheral image and the second peripheral image; when comparing the third comparison module When the difference is the smallest, the determining module uses the average image data for the overlapping portion of the first surrounding image and the second surrounding image; when the comparison difference of the fourth comparing module is the smallest, the determining module is for the first surrounding image and the second surrounding The overlapping part of the image uses the data of the gradation image.

In an embodiment of the invention, the processing unit first corrects the image of the eyeballs to reduce the pincushion distortion, and then combines the corrected images by reducing the pincushion distortion.

The present invention provides a capture method for capturing an image of an eyeball, the capture method comprising simultaneously capturing a plurality of images of the eyeball from a plurality of different directions. The method of capturing also includes merging these images.

In an embodiment of the invention, the images of the eyeballs described above are a plurality of images of the fundus of the eyeball, and the eyeballs are simultaneously captured from the different directions. The steps of these images include capturing these images of the fundus of the eye through the pupil of the eye.

In an embodiment of the invention, adjacent image portions of the fundus of the eyeball are partially overlapped.

In one embodiment of the invention, the step of merging the images includes comparing the overlapping portions of the images as a corrected reference when merging the images.

In an embodiment of the invention, the correction reference includes at least one of a color correction reference, a coordinate conversion correction reference, and a noise cancellation correction reference.

In an embodiment of the invention, the images of the eyeball include a central image and a plurality of peripheral images located adjacent to the central image.

In an embodiment of the invention, the step of merging the images comprises: (a) comparing an overlap between one of the peripheral images and the central image; and (b) taking the peripheral images. Comparing one of the second peripheral images with the overlapping portion of the central image; (c) calculating an average image of the overlapping portion of the first peripheral image and the second surrounding image, and comparing the average image with the overlapping portion of the central image; d) calculating a gradation image of the overlapping portion of the first peripheral image and the second peripheral image, and comparing the overlapping image with the overlapping portion of the central image; and (e) from step (a) to step (d) Taking the step of minimizing the difference after the comparison, when the comparison difference of the step (a) is the smallest, the overlapping portion of the first peripheral image and the second peripheral image adopts the data of the first peripheral image; when the comparison difference of the step (b) is the smallest, The overlapping portion of the first peripheral image and the second peripheral image adopts the number of the second peripheral image According to the data; when the comparison difference of the step (c) is the smallest, the overlapping portion of the first peripheral image and the second peripheral image adopts the average image data; when the comparison difference of the step (d) is the smallest, the first peripheral image and the second periphery The overlapping part of the image uses the data of the gradation image.

In an embodiment of the present invention, the central region of the central image is data of a central image, and the portion of the peripheral region of the central image that overlaps with the central region of the adjacent peripheral images uses data of the peripheral images.

In an embodiment of the invention, the method further includes correcting the pincushion distortions of the images before merging the images, wherein the step of merging the images is to merge by reducing pincushion distortion. These images are corrected.

Based on the above, the image capturing device of the embodiment of the present invention uses a plurality of image sensing modules to respectively capture a plurality of images of the eyeball. Thereby, the time taken for taking an image of the eye multiple times can be reduced, and an eyeball image with a wider viewing angle can be obtained. In the capturing method of the embodiment of the present invention, multiple images of the eyeball can be synchronously captured by different directions, and the images can be combined. By combining these multiple images captured at the same time, the phenomenon of uneven brightness and contrast between multiple images generated by separately capturing the image of the eye can be avoided, thereby improving the efficiency and accuracy of the subsequent merged images.

The above described features and advantages of the present invention will be more apparent from the following description.

1A is a side view of an image capturing device according to an embodiment of the present invention; intention. FIG. 1B is a front elevational view of the image capturing device in accordance with the embodiment of FIG. 1A. Please refer to FIG. 1A and FIG. 1B. In the embodiment, the image capturing device 10 is configured to capture an image of an eyeball 20, and the image capturing device 10 includes a plurality of image sensing modules 100 and at least one light source 200. Each image sensing module 100 includes an image sensor 110 and a lens 120. The image sensor 110 can be a Complementary Metal-Oxide-Semiconductor (CMOS) sensor, a charge coupled device (CCD), or other light sensing suitable for receiving images. Device. Further, in the present embodiment, the lens 120 may be a lens driven by a voice coil motor. In this embodiment, the image capturing device 10 includes five image sensing modules 101, 102, 103, 104, and 105. However, in other embodiments, the image capturing device 10 may include more or The image sensing module 100 is less, and the invention is not limited thereto. Meanwhile, in the embodiment, the image capturing device 10 includes four light sources 200. However, in other embodiments, the number of the light sources 200 and the position of the distribution may be designed according to requirements to achieve the purpose of illumination, and the present invention does not This is limited. The light source 200 may be a light-emitting diode (LED) or other component suitable for emitting light, and the light emitted by the light source 200 may be invisible light such as visible light or infrared light, and the invention is not limited thereto. The light source 200 emits an illumination light L that illuminates the eyeball 20. The eyeball 20 reflects the illumination light L into an image light B. The image light B includes a plurality of sub-image light beams BS (such as the sub-image light beams B1, B2, and B3 in FIG. 1), and the sub-image light beams BS respectively pass through the image sensing modules. These lenses 120 of 100 are passed to these image sensors 110.

In detail, in the present embodiment, the illumination light L can be irradiated to the fundus F of the eyeball 20 via the pupil P of the eyeball 20. The fundus F reflects the illumination light L into the image light B, and the sub-image beams BS of the image light B are respectively transmitted to the image sensing modules 100 via the pupils P. The image capturing range of the fundus F of the adjacent two image sensing modules 100 may partially overlap. Moreover, the optical axes X of the lenses 120 of the image sensing modules 100 may not be parallel to each other, and the optical axes X of the lenses 120 pass through the pupils P of the eyeballs 20 . For example, the image sensing module 101 in FIG. 1A has an optical axis X1, the image sensing module 102 has an optical axis X2, and the image sensing module 103 has an optical axis X3, wherein the optical axes are not parallel to each other. The X1, the optical axis X2, and the optical axis X3 pass through the pupil P, so that the image sensing module 101, the image sensing module 102, and the image sensing module 103 can respectively capture images of different regions on the fundus F from different angles. That is, the image sensing module 101 can capture the image of the fundus region F1, the image sensing module 102 can capture the image of the fundus region F2, and the image sensing module 103 can capture the image of the fundus region F3. The fundus region F1 partially overlaps the fundus region F2, and the fundus region F2 partially overlaps the fundus region F3. Thereby, the image capturing device 10 can simultaneously capture images of different regions of the fundus F, and can assist the medical staff to easily observe the more complete eye image information of the patient, thereby helping the medical staff to diagnose and verify the accuracy and efficiency in the clinical.

In more detail, please refer to FIG. 1A again. In this embodiment, the image sensing module 100 further includes an actuator 130 connectable to at least one of the image sensor 110 and the lens 120. The image sensing module 100 focuses. Wherein, the actuator 130 can be a voice coil motor (voice coil) Motor, VCM) or other type of motor. For example, in the embodiment, the actuator 130 can respectively focus the image sensing modules 101, 102, and 103 on the fundus F of the eyeball 20, because the diopter of the human eye varies from person to person, even the same eye. The diopter of the fundus viewed through the pupil from different angles is also different. Therefore, the focus of each image sensing module 100 is controlled by the actuator 130, and the diopter difference of the eye can be adapted to different human eyes. In turn, it shortens the time of shooting the fundus and increases the image quality.

In addition, the image sensing module 100 can further include a micro processing unit 140 electrically connected to the corresponding image sensor 100 to extract the image data generated by the sub image beam BS measured by the image sensor 100. . The micro processing unit 140 can be a microprocessor such as an image signal processor (ISP). For example, in the embodiment, the image sensing module 101 includes a micro processing unit 141, the image sensing module 102 includes a micro processing unit 142, and the image sensing module 103 includes a micro processing unit 143. That is, each image sensing module 100 can have a respective micro processing unit 140 as a subsystem for fundus image processing. In addition, the image capturing device 100 may further include a processing unit 150 electrically connected to the image sensing modules 100 to respectively generate the plurality of the sub-image beams BS measured by the image sensors 100. The images of the eyeball 20 merge. The processing unit 150 may be a processor such as a digital signal processor (DSP). For example, referring to FIG. 2, in the embodiment, the image sensing modules 101, 102, 103, 104, and 105 respectively have micro processing units 141, 142, 143, 144, and 145 corresponding thereto, and the image sense is The test modules 101, 102, 103, 104 and 105 also have random access memories corresponding thereto Random access memory (RAM) RAM1, RAM2, RAM3, RAM4, and RAM5 can store image information processed by the microprocessing units 141, 142, 143, 144, and 145. In addition, the processing unit 150 may combine the image information processed by the image sensing modules 101, 102, 103, 104, and 105 through the processing unit 140, and may also store the combined result or the data in the operation process. The memory unit SR, wherein the memory unit SR is, for example, a synchronous dynamic random access memory (SDRAM). Therefore, the efficiency of image combining can be effectively increased and the high cost of using a fast processor can be saved. At the same time, in the embodiment, since the micro processing unit 140 is combined with the processing unit 150, the time of fundus shooting can be shortened, so that the fundus images of different regions simultaneously acquired by multiple angles can be continuously processed, and displayed on a display unit DU. The combined fundus image is displayed on the top, which further enables the image capturing device 10 to have a live view function to assist in focusing the fundus image, thereby increasing the quality and accuracy of the fundus image.

For details, please refer to FIG. 1A, FIG. 1B and FIG. In this embodiment, the images of the eyeballs 20 measured by the adjacent two image sensors 110 partially overlap, and the images of the eyeballs 20 may include a central image P0 and a plurality of peripheral images P located beside the central image P0. Processing unit 150 can compare the overlapping portions of these images as a correction reference when merging the images. The calibration reference includes at least one of a color correction reference, a coordinate conversion correction reference, and a noise cancellation correction reference. More specifically, please refer to FIG. 1A, FIG. 1C and FIG. The processing unit 150 can include a first comparison module M1, a second comparison module M2, and a third comparison module M3. A fourth comparison module M4 and a determination module MJ. The first comparison module M1 compares the overlapping portion of the first peripheral image P1 and the central image P0 of the peripheral images P, that is, the data of the first peripheral image P1 in the overlapping region P01 (ie, the region marked with a diagonal line). The data in the overlap region P01 is compared with the central image P0. The second comparison module M2 compares the overlap between the second peripheral image P2 and the central image P1 of the peripheral image P, that is, the data of the second peripheral image P2 in the overlapping area P02 (ie, the area of the cross) The central image P1 is in the data of the overlap region P02. The third comparison module M3 calculates an average image of the overlapping portion of the first peripheral image P1 and the second peripheral image P2, and compares the average image with the overlapping portion of the central image P0, that is, the first peripheral image P1 is in the overlapping region. The data of P12 (i.e., the area in which the oblique line and the cross are drawn) and the data of the second surrounding image P2 in the overlapping area P12 are averaged, and the averaged data is compared with the data of the central image P0 in the overlapping area P12. The fourth comparison module M4 calculates a gradation image of the overlapping portion of the first peripheral image P1 and the second peripheral image P2, and compares the overlapping image with the overlapping portion of the central image P0, that is, the first peripheral image P1 is The data of the overlap region P12 and the data of the second peripheral image P2 in the overlap region P12 are subjected to the gradation image calculation, and the data of the calculation result is compared with the data of the central image P0 in the overlap region P12. The judging module MJ judges which of the comparison results of the first to fourth comparison modules M1, M2, M3, and M4 has the smallest difference. When the comparison difference of the first comparison module M1 is the smallest, the determination module MJ uses the data of the first peripheral image P1 for the overlapping portion of the first peripheral image P1 and the second peripheral image P2. And when the ratio of the second comparison module M2 When the difference is the smallest, the judging module uses the data of the second peripheral image P2 for the overlapping portion of the first peripheral image P1 and the second peripheral image P2. When the comparison difference of the third comparison module M3 is the smallest, the determination module MJ uses the data of the average image for the overlapping portion of the first peripheral image P1 and the second peripheral image P2. When the comparison difference of the fourth comparison module M4 is the smallest, the determination module MJ uses the data of the gradation image for the overlapping portion of the first peripheral image P1 and the second peripheral image P2. Thereby, the plurality of images received by the image sensing module 100 can be spliced in such a manner that the difference between the two is the smallest and the content is the most correct. In general, the central image P0 is an image close to the central region of the fundus F, and the distortion of the image such as pincushion distortion is smaller than that of the image from the sample area of the fundus F, so that it is easy to be corrected later. Taking the central image P0 as a reference image, supplemented by the error of the reference central image P0 and other peripheral images P to merge the images of the fundus F, the accuracy of image stitching can be further increased. The number of the peripheral images P described in FIG. 3 of the present embodiment is only used to illustrate the present invention. Actually, the images participating in the calculation may be different according to the number of images actually captured, and the present invention does not limit.

In addition, in the present embodiment, the central image P0 and the peripheral image P (in the present embodiment are exemplified by the first peripheral image P1, the second peripheral image P2, the third peripheral image P3, and the fourth peripheral image P4, however, in other In the embodiment, the number of peripheral images may be increased or decreased according to actual needs, and the invention is not limited thereto. There are partially overlapping regions. Among them, since the human eye has diopter, the image of the fundus F is usually more pronounced at the outer edge of the image than at the center of the image. In the embodiment, the processing unit 150 can be used for the central image P0. The central area CZ uses the data of the central image P0, and the processing unit 150 uses the data of the peripheral image P for the portion of the peripheral area SZ of the central image P0 that overlaps with the central area CZ of the adjacent peripheral images P. That is, when splicing the image of the fundus F, the central portion of the single fundus image is used as much as possible to avoid the use of the region where the outer edge of the image has a relatively sharp image distortion. In this embodiment, the processing unit 150 may first correct the image of the eyeballs 20 to reduce the pincushion distortion and then combine them. Especially in the outer edge of the image with severe image distortion, the way to correct the pincushion distortion can usually insert additional image points to compensate for the decrease in resolution caused by the correction. Therefore, avoiding the use of the outer edge of the image to merge the image of the fundus F can reduce the image quality degradation caused by the additionally inserted image points.

The first comparison module M1, the second comparison module M2, the third comparison module M3, the fourth comparison module M4, and the determination module MJ may be programs stored in the storage medium of the image capturing device 10, It can be loaded into the processing unit 150 to perform the above functions. Alternatively, in other embodiments, the first comparison module M1, the second comparison module M2, the third comparison module M3, the fourth comparison module M4, and the determination module MJ may also be hard components composed of logic circuit components. The device is configured to perform the functions described above.

4 is a flow chart showing a method of capturing in an embodiment of the present invention. Please refer to FIG. 1A, FIG. 3 and FIG. 4. In this embodiment, the capture method is used to capture an image of an eyeball 20. The capturing method includes simultaneously capturing a plurality of images P of the eyeball 20 from a plurality of different directions (step S10). The capturing method also includes combining the images P (step S20). Wherein, the images P of the eyeball 20 may be a plurality of images P of the fundus F of the eyeball 20, and from these differences The step S10 of simultaneously capturing the images P of the eyeball 20 may include capturing the images P of the fundus F of the eyeball 20 via the pupil P of the eyeball 20. Thereby, the brightness and contrast of the images of the plurality of parts of the fundus F that are simultaneously captured are similar, and the subsequent image combination is easy, which can save the operation time of the image combination and improve the combined image quality, and is convenient for the diagnosis of clinical eye diseases.

The images of the fundus F of the eyeball 20 include a central image P0 and a plurality of peripheral images P located beside the central image P0. Moreover, the step S20 of image merging may include comparing overlapping portions of the images as a correction reference when merging the images. In addition, in this embodiment, the capturing method may further include correcting the pincushion distortion of the images before combining the images (step S10a), wherein the step S20 of combining the images is merged to reduce These images are corrected for pincushion distortion. The effect of the combined image is as described in the embodiment of FIG. 1A, and therefore will not be described herein.

For details, please refer to FIG. 5. The step S20 of combining the images includes: (a) comparing the overlapping portions of the first peripheral image P1 and the central image P0 of the peripheral images P (step S20a); (b) taking one of the peripheral images P Comparing the overlapping portion of the second peripheral image P2 with the central image P0 (step S20b); (c) calculating an average image of the overlapping portion of the first peripheral image P1 and the second peripheral image P2, and taking the average image and the central image P0 The overlapping portion is compared (step S20c); (d) the gradation image of the overlapping portion of the first peripheral image P1 and the second peripheral image P2 is calculated, and the overlapping portion of the gradation image and the central image P0 is compared (step S20d); and (e) the step of taking the smallest difference after the comparison from the step (a) to the step (d) (step S20e). Wherein, in more detail, when the comparison difference of the step (a) is the smallest, the overlapping portion of the first peripheral image P1 and the second peripheral image P2 adopts the data of the first peripheral image P1; when the comparison of the step (b) is the smallest difference The overlapping portion of the first peripheral image P1 and the second peripheral image P2 uses data of the second peripheral image P2; when the comparison difference of the step (c) is the smallest, the overlapping portion of the first peripheral image P1 and the second peripheral image P2 The data of the average image is used; when the comparison difference of the step (d) is the smallest, the overlapping portion of the first peripheral image P1 and the second peripheral image P2 uses the data of the progressive image. The central area CZ of the central image P0 uses the data of the central image P0, and the portion of the peripheral area SZ of the central image P0 that overlaps with the central area CZ of the adjacent peripheral image P uses the data of the peripheral image P. A more detailed process of combining images and its efficacy is illustrated in the embodiment of Figure 1A. The steps (a) to (e) can be performed by using the first comparison module M1, the second comparison module M2, the third comparison module M3, the fourth comparison module M4, and the determination module MJ. For details, refer to the description of the functions performed by these modules, and therefore will not be described here. Moreover, the order of the above steps is an example of the invention, and the invention is not limited thereto.

For example, please refer to Figure 6. In this embodiment, the process of capturing the fundus image may further include automatic pupil P detection (step S5) and determining whether the pupil P is detected (step S6), if the image of the pupil P is not detected, repeating Step S5 of detecting the execution of the pupil P detection is detected. If the image of the pupil P has been detected, the plurality of images of the eyeball 20 can be simultaneously captured from a plurality of different directions (step S10), and each lens 120 is driven to focus (step S11). and Thereafter, each lens 120 on the image sensing module 100 (for example, N lenses 120, such as lens 1, lens 2, ..., and lens N, where N is a positive integer greater than 1) can be separately focused (step S12) And it is judged whether or not the focus is successful (step S13). If the focus is not successful, the process returns to step S12 to focus again. When all the lenses 120 (including the lens 120 located at the central portion of the image sensing device 10 and the surrounding portion) are in focus (step S14), the image sensing module 100 is driven to take a picture (step S15). The image sensing module 100 and the lens 120 can take a picture and capture an image of the fundus F (step S16). Then, these images are combined by the processing unit 150 (step S20) and these fundus images are output (step S30). Thereby, the image sensing device 10 can automatically detect human eye physiological information, and output a wide range of clear fundus images for medical staff diagnosis.

In summary, the image capturing device of the embodiment of the present invention can be used to simultaneously capture multiple fundus images of different parts of the eyeball. Since these fundus images can be captured at the same time, the brightness and contrast are similar. Then, by comparing the difference between the overlapping portions of the fundus images by the processing unit and combining the fundus images with the smallest difference between the images, the time required for the combined images can be effectively saved, and the wide range and image quality can be quickly obtained. Good fundus imagery. In addition, each image sensing module can include an actuator and can simultaneously focus on different regions of the fundus at different angles, which can save time for taking fundus images, reduce the burden on the patient's eyes, and increase the clear fundus of a wide range of images. The success rate of the image, which in turn improves the quality of the medical care and the diagnostic accuracy of the medical staff.

Although the invention has been disclosed above by way of example, it is not intended to be limiting The scope of the present invention is defined by the scope of the appended claims, and the scope of the invention is defined by the scope of the appended claims. Prevail.

10‧‧‧Image capture device

20‧‧‧ eyeballs

100, 101, 102, 103, 104, 105‧‧‧ image sensing module

200‧‧‧Light source

110‧‧‧Image Sensor

120‧‧‧ lens

L‧‧‧Lights

B‧‧‧Image light

BS, B1, B2, B3‧‧‧ sub-image beam

F‧‧‧ fundus

X, X1, X2, X3‧‧‧ optical axis

P‧‧‧ pupil

F1, F2, F3‧‧‧ fundus area

130‧‧‧Actuator

140, 141, 142, 143, 144, 145‧‧‧ microprocessing units

150‧‧‧Processing unit

DU‧‧‧ display unit

RAM1, RAM2, RAM3, RAM4, RAM5‧‧‧ random access memory

SR‧‧‧ memory unit

P0‧‧‧ central image

P‧‧‧ peripheral image

M1‧‧‧ first comparison module

M2‧‧‧Second comparison module

M3‧‧‧ third comparison module

M4‧‧‧fourth comparison module

MJ‧‧‧ judgment module

CZ‧‧‧Central District

SZ‧‧‧ surrounding area

S5, S6, S10, S10a, S11, S12, S13, S14, S15, S16, S20, S20a, S20b, S20c, S20d, S20e, S30‧‧

1A is a side elevational view of an image capture device in accordance with an embodiment of the present invention.

FIG. 1B is a front elevational view of the image capturing device in accordance with the embodiment of FIG. 1A.

Figure 1C is a schematic illustration of a processing unit in accordance with the embodiment of Figure 1A.

2 is a block diagram of an image capture device in accordance with the embodiment of FIG. 1A.

3 is a schematic view of a plurality of fundus images superimposed in accordance with the embodiment of FIG. 1A.

FIG. 4 is a schematic flow chart of a method for capturing in accordance with the embodiment of FIG. 1A.

Figure 5 is a flow chart showing the detailed steps in accordance with step S20 of Figure 4.

Figure 6 is a flow chart of capturing a fundus image in accordance with the embodiment of Figure 4.

10‧‧‧Image capture device

20‧‧‧ eyeballs

100‧‧‧Image Sensing Module

200‧‧‧Light source

110‧‧‧Image Sensor

120‧‧‧ lens

101, 102, 103‧‧‧ image sensing module

L‧‧‧Lights

B‧‧‧Image light

BS, B1, B2, B3‧‧‧ sub-image beam

F‧‧‧ fundus

X, X1, X2, X3‧‧‧ optical axis

P‧‧‧ pupil

F1, F2, F3‧‧‧ fundus area

130‧‧‧Actuator

140, 141, 142, 143‧‧‧ microprocessing units

150‧‧‧Processing unit

Claims (23)

An image capturing device for capturing an image of an eyeball, the image capturing device comprising: a plurality of image sensing modules, each of the image sensing modules comprising an image sensor and a lens; and at least a light source that emits an illumination light that is incident on the eyeball, the eyeball reflecting the illumination light into an image light, the image light comprising a plurality of sub-image light beams, wherein the sub-image light beams respectively pass through the image sensing modules The lenses are transmitted to the image sensors. The image capturing device of claim 1, wherein the illumination light is incident on a fundus of the eyeball through a pupil of the eyeball, and the fundus reflects the illumination light into the image light, and the image light The sub-image beams are respectively transmitted to the image sensing modules via the pupils. The image capturing device of claim 2, wherein two adjacent image sensing modules partially overlap the image capturing range of the fundus. The image capturing device of claim 1, wherein the optical axes of the lenses of the image sensing modules are not parallel to each other, and the optical axes of the lenses pass through the pupil of the eyeball. The image capturing device of claim 1, wherein each of the image sensing modules further includes an actuator coupled to at least one of the image sensor and the lens to enable the image The sensing module focuses. The image capturing device of claim 5, wherein each of the image sensing modules further includes a micro processing unit electrically connected to the corresponding image sensor to take out the image sensor Measured Information about the image produced by the beam. The image capturing device of claim 1, further comprising a processing unit electrically connected to the image sensing modules to measure the sub-image beams measured by the image sensors The images of the plurality of eyeballs generated separately are combined. The image capturing device of claim 7, wherein the images of the eyeballs measured by the adjacent two image sensors partially overlap. The image capturing device of claim 8, wherein the processing unit compares overlapping portions of the images as a correction reference when combining the images. The image capturing device of claim 9, wherein the correction reference comprises at least one of a color correction reference, a coordinate conversion correction reference, and a noise suppression correction reference. The image capturing device of claim 8, wherein the images of the eyeball comprise a central image and a plurality of peripheral images located adjacent to the central image. The image capture device of claim 11, wherein the processing unit comprises: a first comparison module, wherein an overlap between the first peripheral image and the central image of the peripheral images is taken; Comparing; a second comparison module compares a second peripheral image of the peripheral images with an overlap of the central image; and a third comparison module calculates the first peripheral image and the second week An average image of the overlapping portion of the side image, and comparing the average image with the overlapping portion of the central image; a fourth comparing module calculates a gradient of the overlapping portion of the first surrounding image and the second surrounding image Comparing the image with the overlapping portion of the gradation image and the central image; and determining a comparison module for determining a difference between the comparison results of the first to fourth comparison modules, wherein when the first comparison module is When the comparison difference is the smallest, the determining module uses the data of the first peripheral image for the overlapping portion of the first peripheral image and the second peripheral image; and when the comparison difference of the second comparison module is the smallest, the determining module The data of the second peripheral image is used for the overlapping portion of the first peripheral image and the second peripheral image; and when the comparison difference of the third comparison module is the smallest, the determining module is configured for the first surrounding image and the first The overlapping portion of the two peripheral images adopts the data of the average image; when the comparison difference of the fourth comparison module is the smallest, the determining module is configured for the first peripheral image and the second Using overlapping portion of video image data of the image layer gradually. The image capturing device of claim 11, wherein the processing unit uses the data of the central image for the central region of the central image, and the processing unit is adjacent to the adjacent region of the central image. The overlapping portions of the peripheral images of the peripheral images use the data of the peripheral images. The image capturing device of claim 7, wherein the processing unit first corrects the image of the eyeballs to reduce the pincushion distortion, and then combines the corrected images by reducing the pincushion distortion. A method for capturing an image of an eyeball, the method comprising: capturing a plurality of images of the eyeball from a plurality of different directions; and combining the images. The method of claim 15, wherein the image of the eyeball is a plurality of images of the fundus of the eyeball, and the step of synchronously capturing the images of the eyeball from the different directions comprises: The images of the fundus of the eyeball are captured through the pupil of the eyeball. The method of claim 16, wherein the two adjacent images of the fundus of the eyeball partially overlap. The method of claim 17, wherein the step of combining the images comprises comparing overlapping portions of the images as a correction reference when combining the images. The method of claim 18, wherein the calibration reference comprises at least one of a color correction reference, a coordinate conversion correction reference, and a noise cancellation correction reference. The method of claim 17, wherein the images of the fundus of the eyeball comprise a central image and a plurality of peripheral images located adjacent to the central image. The method of claim 20, wherein the step of combining the images comprises: (a) comparing one of the peripheral images with the overlap of the central image; b) taking one of the peripheral images and the second peripheral image Comparing the overlapping portions with the image; (c) calculating an average image of the overlapping portion of the first peripheral image and the second surrounding image, and comparing the average image with the overlapping portion of the central image; (d) calculating a gradation image of the overlapping portion of the first peripheral image and the second peripheral image, and comparing the overlapping image with the overlapping portion of the central image; and (e) from step (a) to step (d) Taking the step of the smallest difference after the comparison, when the comparison difference of the step (a) is the smallest, the overlapping portion of the first peripheral image and the second peripheral image adopts the data of the first peripheral image; when the comparison of the step (b) is different If the difference between the first peripheral image and the second peripheral image is the smallest, the data of the second peripheral image is used; when the comparison difference of the step (c) is the smallest, the overlapping of the first peripheral image and the second surrounding image The data of the average image is used in part; when the comparison difference of the step (d) is the smallest, the overlapping portion of the first peripheral image and the second peripheral image adopts the data of the gradation image. The method of claim 20, wherein the central region of the central image uses data of the central image, and a portion of the peripheral region of the central image that overlaps with a central region of the adjacent peripheral images , using the data of the peripheral images. The method of claim 15 further includes: correcting the pincushion distortion of the images before merging the images, wherein the step of merging the images is to reduce by combining The corrected images of the pincushion distortion.