CN107424128B - Method and system for extracting clear image frame from recorded eye fundus video - Google Patents
Method and system for extracting clear image frame from recorded eye fundus video Download PDFInfo
- Publication number
- CN107424128B CN107424128B CN201710467748.6A CN201710467748A CN107424128B CN 107424128 B CN107424128 B CN 107424128B CN 201710467748 A CN201710467748 A CN 201710467748A CN 107424128 B CN107424128 B CN 107424128B
- Authority
- CN
- China
- Prior art keywords
- pixel point
- value
- frame
- image
- qualified
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 23
- 238000012216 screening Methods 0.000 claims description 13
- 210000001747 pupil Anatomy 0.000 abstract description 8
- 238000002360 preparation method Methods 0.000 abstract description 3
- 230000008569 process Effects 0.000 abstract description 3
- 238000003759 clinical diagnosis Methods 0.000 abstract description 2
- 230000007704 transition Effects 0.000 abstract description 2
- 210000001508 eye Anatomy 0.000 description 7
- 238000004364 calculation method Methods 0.000 description 5
- 201000010099 disease Diseases 0.000 description 5
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 5
- 238000003384 imaging method Methods 0.000 description 5
- 230000003902 lesion Effects 0.000 description 3
- 230000001575 pathological effect Effects 0.000 description 3
- 210000001525 retina Anatomy 0.000 description 3
- 210000003161 choroid Anatomy 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 206010000234 Abortion spontaneous Diseases 0.000 description 1
- 206010025421 Macule Diseases 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000017531 blood circulation Effects 0.000 description 1
- 210000001775 bruch membrane Anatomy 0.000 description 1
- 210000005252 bulbus oculi Anatomy 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 230000004089 microcirculation Effects 0.000 description 1
- 208000015994 miscarriage Diseases 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 231100000915 pathological change Toxicity 0.000 description 1
- 230000036285 pathological change Effects 0.000 description 1
- 238000003825 pressing Methods 0.000 description 1
- 208000000995 spontaneous abortion Diseases 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/73—Deblurring; Sharpening
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/80—Geometric correction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30041—Eye; Retina; Ophthalmic
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Medical Informatics (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Eye Examination Apparatus (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a method and a system for extracting clear image frames from a recorded fundus video. The method and system of the present invention can be applied to various fundus image capturing apparatuses. The acquirer records a video directly captured by the camera device in the acquisition process, such as: in the preparation stage, the aim of aligning the entrance pupil of the acquisition equipment with the pupil of the person to be acquired is tried, the acquisition is carried out for several times, high-quality fundus images are captured for a short time or continuously, the acquisition interval is coordinated with the person to be acquired to adjust the shooting angle and the like, and the transition stage after the acquisition is finished. The acquirer can obtain a high-quality fundus image after software processing without manually editing the video or setting software parameters, and manually screen out representative parts for clinical diagnosis.
Description
Technical Field
The invention relates to the field of pathological image processing, in particular to a method and a system for extracting clear image frames from recorded eyeground videos.
Background
The inner part of the eye, i.e. the inner lining of the eyeball, which can be seen through the pupil, includes the retina, optic disc and macula. Due to the transparency of the retina, this term also includes Bruch's membrane and choroid. The fundus is the only part of the human body that can directly see the microcirculation of the blood circulation. Diseases of the fundus, including diseases of the retina, optic disc and choroid, are commonly referred to as ocular fundus diseases. Many ocular fundus diseases are closely related to changes in the vitreous, and therefore, vitreous lesions are also included.
At present, doctors observe the pathological changes of the eyeground of patients through the traditional direct ophthalmoscope, including an illumination system and an observation system, the basic principle of the system is that an ophthalmoscope is used to project light rays to the eyes of the examinees through the pupils of the eyes, and then the light rays reflected by the eyeground of the examinees are imaged in the eyes of the examinees. The traditional direct ophthalmoscope does not have the shooting and recording functions, so that different doctors have high subjectivity in observing and judging the fundus diseases of patients, and visual medical record management cannot be formed. If the fundus image needs to be taken by a fixed instrument, the photographing is very inconvenient for special patients.
The applicant previously filed patent application numbers: 201710240648.X, which is a Chinese patent application, provides an eyeground shooting frame which is convenient to disassemble and assemble and can be quickly connected with a mobile phone, and the mobile phone can clearly shoot the eyeground condition under the assistance of the eyeground shooting frame. In practical use, when one hand holds the shooting frame and the other hand clicks the shooting key, the shooting frame is slightly shaken, and therefore shooting imaging quality is affected.
The above problems need to be solved.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: a method and system for extracting a clear image frame from a recorded fundus video is provided.
The invention provides a technical scheme (one) for solving the technical problems, which comprises the following steps: a method of extracting a clear image frame from a recorded fundus video, comprising the steps of:
converting each pixel point in the effective image area in each frame of image in the recorded video file into a description form of YUV color gamut three channels, and screening out qualified pixel points;
selecting a preset number of adjacent pixel points adjacent to the qualified pixel points in the circumferential direction, and respectively calculating the maximum absolute value of the Y channel value difference value of each qualified pixel point and the adjacent pixel points;
calculating the sum of the maximum absolute values of all qualified pixel points in each frame of image, and dividing the sum by the number of all qualified pixel points in the frame of image to obtain a difference average value;
comparing the average value of the difference values of all the frame images in the recorded video file;
and selecting and outputting image data information with the largest difference average value of a certain frame according to the comparison result.
Further, the step of screening qualified pixel points includes
Judging whether Y, U and V channel values of each pixel point in an effective image area of each frame of image in the recorded video file respectively fall into a preset Y, U value interval and a preset V value interval, and if Y, U and V channel values of a certain pixel point fall into Y, U value interval and the V value interval, determining the pixel point as a qualified pixel point.
Further, when a certain pixel point is judged, whether the Y channel value of the pixel point falls into a preset Y value range is judged, if the Y channel value of the pixel point falls into the preset Y value range, the pixel point is judged to be a primary qualified pixel point, and if the Y channel value of the pixel point does not fall into the preset Y value range, the pixel point is judged to be an invalid pixel point;
and judging whether the U channel value and the V channel value of the preliminary qualified pixel point both fall into a preset U value and V value interval, and if so, determining the pixel point as a final qualified pixel point.
Furthermore, a preset number of adjacent pixels of a qualified pixel are extracted from all pixels closely adjacent to the qualified pixel according to a preset rule.
Further, before the screening, the recorded video file is divided into n segments according to a preset time interval;
when the difference average value is compared, the difference average value of all the frame images in each video segment is separately compared;
when outputting, respectively outputting the image with the maximum difference average value of a certain frame in each video segment, and outputting n frames of image data information.
The invention provides a technical scheme (II) for solving the technical problems, which comprises the following steps: a system for extracting clear image frames from recorded fundus video includes
A video file storage unit suitable for storing image data information of each frame of recorded video
A buffer unit;
a processor unit adapted to
Sequentially reading the image data information of each frame from the video file storage unit according to the time axis of the recorded video file, and performing the reading
Converting each pixel point in the effective image area in the current frame image into a three-channel description form of YUV color gamut, screening out qualified pixel points,
determining a preset number of adjacent pixel points adjacent to the qualified pixel points in the circumferential direction of each qualified pixel point, respectively calculating the maximum absolute value of the Y channel value difference between each qualified pixel point and the adjacent pixel point,
calculating the sum of the maximum absolute values of all qualified pixel points in the current frame image, dividing the sum by the number of all qualified pixel points in the frame image to obtain a difference average value,
selecting to store the current frame image data information into the cache unit or release the current frame image data information as invalid data according to a preset rule;
and the output delivery unit is suitable for outputting the one or more frames of images stored in the cache unit to an external display device.
Further, the processor unit is used for screening qualified pixel points
And judging whether Y, U and the V channel value of each pixel point in the effective image area of the current frame image respectively fall into a preset Y, U value interval and a preset V value interval, and if Y, U and the V channel value of a certain pixel point both fall into a Y, U value interval and a V value interval, determining the pixel point as a qualified pixel point.
Further, when a certain pixel point is judged, whether the Y channel value of the pixel point falls into a preset Y value range is judged, if the Y channel value of the pixel point falls into the preset Y value range, the pixel point is judged to be a primary qualified pixel point, and if the Y channel value of the pixel point does not fall into the preset Y value range, the pixel point is judged to be an invalid pixel point;
and judging whether the U channel value and the V channel value of the preliminary qualified pixel point both fall into a preset U value and V value interval, and if so, determining the pixel point as a final qualified pixel point.
The system further comprises a video dividing unit which is suitable for dividing the recorded video file into n video files according to a preset time interval;
the video file storage unit is divided into n logic storage subunits corresponding to the n sections of video files, and each logic storage subunit is suitable for correspondingly storing each frame of image data information of each divided section of video file;
the cache unit is divided into n logic cache subunits corresponding to the n video files;
the processor unit is suitable for selecting the current frame image data information to be stored in one of the n logic cache subunits corresponding to the current frame image data information or to be released as invalid data according to a preset rule.
Further, the preset rule followed by the processor unit when selecting the release or the release comprises
When the current frame is the first frame image of any section of video file, the image data information of the current frame is selected to be combined with the difference average value information of the frame and then put into the cache unit and the corresponding logic cache subunit;
when the current frame is not the first frame image of any video file, judging whether the difference average value of the current frame is larger than the difference average value of the image frames in the corresponding logic cache subunit;
if the difference average value of the current frame is judged to be larger than the difference average value of the existing image frame, deleting the image data information of the existing image frame in the logic buffer subunit, and putting the image data information of the current frame into the logic buffer subunit by combining the difference average value with the image data information of the current frame
And releasing the current frame image data information as invalid data when the difference average value of the current frame is judged to be less than or equal to the difference average value of the existing image frame.
The invention has the beneficial effects that:
the method and the system for extracting the clear image frame from the recorded fundus video can be applied to various fundus image acquisition devices. The acquirer uses the device to record a video directly captured by the camera device in the acquisition process, wherein the method can comprise the following steps: in the preparation stage, the aim of aligning the entrance pupil of the acquisition equipment with the pupil of the acquired person is tried, the acquisition is carried out for several times, high-quality fundus images are captured for a short time or continuously, the shooting angle is adjusted in the acquisition gap in a coordinated mode with the acquired person, and the like, and the transition stage after the acquisition is finished is also tried. The acquirer can obtain a high-quality fundus image after software processing without manually editing the video or setting software parameters, and manually screen out representative parts for clinical diagnosis.
The method for extracting the clear image frame from the recorded fundus video has the following advantages:
1. when the shutter is used for triggering the static image capture, the limb action of an acquirer causes the optical path entrance pupil of the acquisition equipment and the eye system exit pupil of an acquirer to lose accurate position matching, so that the imaging quality is reduced;
2. the situation that the delay from the short-time accurate matching to the shutter pressing of the acquirer is missed by the best matching time window under the visual observation of the acquirer, so that the imaging quality is reduced or the imaging chance is lost can be avoided;
3. manual intervention of an acquirer in the acquisition process is reduced, attention does not need to be paid to trigger a shutter, and the position of equipment is better focused on adjusting and the imaging quality is improved. For portable equipment, particularly handheld fundus image acquisition equipment, the invention can obviously improve the smoothness of operation and reduce the difficulty of use.
Detailed Description
Example one
The method for extracting clear image frames from the recorded fundus video needs to prepare pathological videos about the bottoms of the eyes of patients, which are recorded by a shooting device, before implementation, and the pathological video files are called recorded video files.
When the method is implemented, the effective image area of each frame of image in the recorded video file is determined, and the scheme for determining the effective image area can be any one of the prior art. For example, by using a shadow tube or by correcting for distortion, a circular central image region and a black border around the perimeter of the circular central image region are formed in each frame of image, and the effective image region can be determined by applying a sophisticated algorithm to disregard or ablate the black border. Alternatively, the effective image area may be determined by shrinking a predetermined number of pixels inward from the edge of each original image.
The image is initially processed for eliminating the miscarriage impurities and correcting the edge distortion, and many mature methods are available in the prior art, and the key point of the invention is not in this, so the description is omitted.
And then converting each pixel point in the effective image area in each frame of image in the recorded video file into a description form of YUV color gamut three channels. Also, the format conversion method employs the prior art scheme.
Qualified pixel points are further screened out, and qualified specific pixel points of the qualified pixel points are worth calculating and have specific reference significance, but are not qualified in popular significance, and the following is the same. The "qualified" in the colloquial meaning may be that a certain pixel has a color or contains a certain amount of information.
In the judgment, the preferred scheme adopted in the embodiment is as follows: and (3) inspecting whether Y, U and the V channel value of each pixel point in the effective image area respectively fall into a preset Y, U value interval and a preset V value interval, and if Y, U and the V channel value of a certain pixel point both fall into a Y, U value interval and the V value interval, determining the pixel point as a qualified pixel point.
To be able to further reduce the computational effort, it may be preferable to: when a certain pixel point is judged, whether the Y channel value of the pixel point falls into a preset Y value range or not is judged, if the Y channel value of the pixel point falls into the preset Y value range, the pixel point is judged to be a primary qualified pixel point, if the Y channel value does not fall into the preset Y value range, the pixel point is judged to be an invalid pixel point, and the invalid pixel point is not subjected to next-step calculation, so that the calculation amount is reduced.
And judging whether the U channel value and the V channel value of the primary qualified pixel point both fall into a preset U value and V value range, and if so, determining the pixel point as a final qualified pixel point.
And then selecting five adjacent pixel points adjacent to the five pixel points in the circumferential direction of each qualified pixel point, and respectively calculating the maximum absolute value of the Y channel value difference value of each qualified pixel point and the adjacent pixel point. Of course, five of the embodiments are for the consideration of the computing capability of the existing computing chip, and four, six, etc. may be selected.
In this embodiment, five adjacent pixels of a qualified pixel are extracted from all pixels immediately adjacent to the qualified pixel according to a predetermined rule. The preset rule is to select five pixel points on the left, right and lower sides of the qualified pixel point. The preset rule can be flexibly changed according to the requirements of different occasions, such as selecting five pixel points on the left, right and upper sides of the qualified pixel point.
And calculating the sum of the maximum absolute values of all qualified pixel points in each frame of image, and dividing the sum by the number of all qualified pixel points in the frame of image to obtain the average value of the difference values.
And comparing the average value of the difference values of all the frame images in the recorded video file.
And selecting and outputting image data information with the maximum difference average value of a certain frame according to the comparison result. And if the average values of the differences of two encountered frames are the same and are the maximum, preferentially outputting the frame closest to the tail end of the time axis of the video file.
In order to more fully reveal the condition of the eyeground and avoid the output clearest image without being related to the eyeground part of a patient. The recorded video file may be divided into n segments at preset time intervals before the screening.
When the difference average value size is compared, the difference average value sizes of all the frame images in each video are compared separately.
When outputting, respectively outputting the image with the maximum difference average value of a certain frame in each video segment, and totally outputting n frames of image data information.
This is avoided by the preferred embodiment even if the recorded video file contains preparations, approaching movements near the lesion just started to be recorded and/or movements away from the lesion near the end of the recording.
Example two
The present embodiment provides a system for extracting a clear image frame from a recorded fundus video, the system embodying or facilitating implementation of the method of the first embodiment, comprising:
and the video file storage unit is suitable for storing the image data information of each frame of the recorded video.
A buffer unit.
A processor unit adapted to perform the following operations and calculations:
sequentially reading the image data information of each frame from the video file storage unit according to the time axis of the recorded video file, converting each pixel point in the effective image area in the current frame image into a description form of YUV color gamut three-channel after reading, and screening out qualified pixel points. When the specific processor unit is used for screening qualified pixel points, whether Y, U and a V channel value of each pixel point in an effective image area of a current frame image respectively fall into a preset Y, U value interval and a preset V value interval is judged, and if the Y, U and the V channel value of a certain pixel point both fall into the Y, U value interval and the V value interval, the pixel point is determined to be a qualified pixel point. The selection of the effective image area and the qualified pixel points are significant in the invention and are related to the first embodiment, and thus the details are not repeated. To be able to further reduce the computational effort, it may be preferable to: when a certain pixel point is judged, whether the Y channel value of the pixel point falls into a preset Y value range or not is judged, if the Y channel value of the pixel point falls into the preset Y value range, the pixel point is judged to be a primary qualified pixel point, if the Y channel value does not fall into the preset Y value range, the pixel point is judged to be an invalid pixel point, and the invalid pixel point is not subjected to next-step calculation, so that the calculation amount is reduced. And judging whether the U channel value and the V channel value of the primary qualified pixel point both fall into a preset U value and V value range, and if so, determining the pixel point as a final qualified pixel point.
And
and selecting five adjacent pixel points adjacent to the qualified pixel point in the circumferential direction, and respectively calculating the maximum absolute value of the Y channel value difference value of each qualified pixel point and the adjacent pixel point. Of course, five of the embodiments are for the consideration of the computing capability of the existing computing chip, and four, six, etc. may be selected.
And
and calculating the sum of the maximum absolute values of all qualified pixel points in the current frame image, and dividing the sum by the number of all qualified pixel points in the frame image to obtain a difference average value.
And
and selecting to store the current frame image data information into a cache unit or release the current frame image data information as invalid data according to a preset rule.
The device also comprises an output delivery unit which is suitable for outputting the one or more frames of images stored in the buffer unit to an external display device.
In order to more fully reveal the condition of the eyeground and avoid the output clearest image without being related to the eyeground part of a patient. Preferred may be: the video dividing unit is suitable for dividing the recorded video file into n video files according to the preset time interval. The video file storage unit is also divided into n logic storage subunits corresponding to n video files, and each logic storage subunit is suitable for correspondingly storing each frame of image data information of each divided video file. The cache unit is divided into n logic cache subunits corresponding to n video files. The processor unit is suitable for selecting the current frame image data information to be stored in one of the n corresponding logic buffer subunits according to a preset rule or to be released as invalid data. The preset rule of the processor unit in selecting release may include that when the current frame is the first frame image of any section of video file, the processor unit selects to release the image data information of the current frame in combination with the difference average value information of the frame into the cache unit and the corresponding logic cache subunit; when the current frame is not the first frame image of any video file, judging whether the difference average value of the current frame is larger than the difference average value of the image frames in the corresponding logic cache subunit; if the difference average value of the current frame is judged to be larger than the difference average value of the existing image frame, deleting the image data information of the existing image frame in the logic buffer subunit, and putting the image data information of the current frame into the logic buffer subunit in combination with the difference average value thereof, and if the difference average value of the current frame is judged to be smaller than or equal to the difference average value of the existing image frame, releasing the image data information of the current frame as invalid data.
The present invention is not limited to the above embodiments, and the technical solutions of the above embodiments of the present invention may be combined with each other in a crossing manner to form a new technical solution, and all technical solutions formed by using equivalent substitutions fall within the scope of the present invention.
Claims (10)
1. A method for extracting clear image frames from recorded fundus videos comprises the following characteristic steps:
converting each pixel point in the effective image area in each frame of image in the recorded video file into a description form of YUV color gamut three channels, and screening out qualified pixel points;
selecting a preset number of adjacent pixel points adjacent to the qualified pixel points in the circumferential direction, and respectively calculating the maximum absolute value of the Y channel value difference value of each qualified pixel point and the adjacent pixel points;
calculating the sum of the maximum absolute values of all qualified pixel points in each frame of image, and dividing the sum by the number of all qualified pixel points in the frame of image to obtain a difference average value;
comparing the average value of the difference values of all the frame images in the recorded video file;
and selecting and outputting image data information with the largest difference average value of a certain frame according to the comparison result.
2. A method of extracting clear image frames from a recorded fundus video according to claim 1, wherein: the step of screening qualified pixel points comprises
Judging whether Y, U and V channel values of each pixel point in an effective image area of each frame of image in the recorded video file respectively fall into a preset Y, U value interval and a preset V value interval, and if Y, U and V channel values of a certain pixel point fall into Y, U value interval and the V value interval, determining the pixel point as a qualified pixel point.
3. A method of extracting clear image frames from a recorded fundus video according to claim 2, wherein: when a certain pixel point is judged, firstly judging whether the Y channel value of the pixel point falls into a preset Y value range, if so, judging the pixel point to be a primary qualified pixel point, and if not, judging the pixel point to be an invalid pixel point;
and judging whether the U channel value and the V channel value of the preliminary qualified pixel point both fall into a preset U value and V value interval, and if so, determining the pixel point as a final qualified pixel point.
4. A method of extracting clear image frames from a recorded fundus video according to claim 1, wherein: and capturing a preset number of adjacent pixels of a qualified pixel from all pixels which are closely adjacent to the qualified pixel according to a preset rule.
5. A method of extracting clear image frames from a recorded fundus video according to claim 1, wherein:
before the screening, dividing the recorded video file into n sections according to a preset time interval;
when the difference average value is compared, the difference average value of all the frame images in each video segment is separately compared;
when outputting, respectively outputting the image with the maximum difference average value of a certain frame in each video segment, and outputting n frames of image data information.
6. A system for extracting a clear image frame from a recorded fundus video, comprising: comprises that
A video file storage unit suitable for storing image data information of each frame of recorded video
A buffer unit;
a processor unit adapted to
Sequentially reading the image data information of each frame from the video file storage unit according to the time axis of the recorded video file, and performing the reading
Converting each pixel point in the effective image area in the current frame image into a three-channel description form of YUV color gamut, screening out qualified pixel points,
determining a preset number of adjacent pixel points adjacent to the qualified pixel points in the circumferential direction of each qualified pixel point, respectively calculating the maximum absolute value of the Y channel value difference between each qualified pixel point and the adjacent pixel point,
calculating the sum of the maximum absolute values of all qualified pixel points in the current frame image, dividing the sum by the number of all qualified pixel points in the frame image to obtain a difference average value,
selecting to store the current frame image data information into the cache unit or release the current frame image data information as invalid data according to a preset rule;
and the output delivery unit is suitable for outputting the one or more frames of images stored in the cache unit to an external display device.
7. The system for extracting clear image frames from recorded fundus video of claim 6, wherein: the processor unit is used for screening qualified pixel points
And judging whether Y, U and the V channel value of each pixel point in the effective image area of the current frame image respectively fall into a preset Y, U value interval and a preset V value interval, and if Y, U and the V channel value of a certain pixel point both fall into a Y, U value interval and a V value interval, determining the pixel point as a qualified pixel point.
8. The system for extracting clear image frames from recorded fundus video of claim 7, wherein: when a certain pixel point is judged, firstly judging whether the Y channel value of the pixel point falls into a preset Y value range, if so, judging the pixel point to be a primary qualified pixel point, and if not, judging the pixel point to be an invalid pixel point;
and judging whether the U channel value and the V channel value of the preliminary qualified pixel point both fall into a preset U value and V value interval, and if so, determining the pixel point as a final qualified pixel point.
9. A system for extracting clear image frames from recorded fundus video according to claim 8, wherein: the video dividing unit is suitable for dividing the recorded video file into n video files according to a preset time interval;
the video file storage unit is divided into n logic storage subunits corresponding to the n sections of video files, and each logic storage subunit is suitable for correspondingly storing each frame of image data information of each divided section of video file;
the cache unit is divided into n logic cache subunits corresponding to the n video files;
the processor unit is suitable for selecting the current frame image data information to be stored in one of the n logic cache subunits corresponding to the current frame image data information or to be released as invalid data according to a preset rule.
10. A system for extracting clear image frames from recorded fundus video according to claim 9, wherein: the processor unit follows a predetermined rule when selecting the release or the release including
When the current frame is the first frame image of any section of video file, the image data information of the current frame is selected to be combined with the difference average value information of the frame and then put into the cache unit and the corresponding logic cache subunit;
when the current frame is not the first frame image of any video file, judging whether the difference average value of the current frame is larger than the difference average value of the image frames in the corresponding logic cache subunit;
if the difference average value of the current frame is judged to be larger than the difference average value of the existing image frame, deleting the image data information of the existing image frame in the logic buffer subunit, and putting the image data information of the current frame into the logic buffer subunit by combining the difference average value with the image data information of the current frame
And releasing the current frame image data information as invalid data when the difference average value of the current frame is judged to be less than or equal to the difference average value of the existing image frame.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710467748.6A CN107424128B (en) | 2017-06-20 | 2017-06-20 | Method and system for extracting clear image frame from recorded eye fundus video |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710467748.6A CN107424128B (en) | 2017-06-20 | 2017-06-20 | Method and system for extracting clear image frame from recorded eye fundus video |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107424128A CN107424128A (en) | 2017-12-01 |
| CN107424128B true CN107424128B (en) | 2020-12-29 |
Family
ID=60426094
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710467748.6A Active CN107424128B (en) | 2017-06-20 | 2017-06-20 | Method and system for extracting clear image frame from recorded eye fundus video |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107424128B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110840430B (en) * | 2018-08-21 | 2022-09-13 | 北京万生人和科技有限公司 | Intra-abdominal pressure data screening method, computer-readable storage medium, and intra-abdominal pressure data screening device |
| CN109658395B (en) * | 2018-12-06 | 2022-09-09 | 代黎明 | Optic disc tracking method and system and eye fundus collection device |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103501406A (en) * | 2013-09-16 | 2014-01-08 | 北京智谷睿拓技术服务有限公司 | Image collecting system and image collecting method |
| CN105049875A (en) * | 2015-07-24 | 2015-11-11 | 上海上大海润信息***有限公司 | Accurate key frame extraction method based on mixed features and sudden change detection |
| CN105049734A (en) * | 2015-08-31 | 2015-11-11 | 广州市幸福网络技术有限公司 | License camera capable of giving shooting environment shooting prompt and shooting environment detection method |
| CN105187728A (en) * | 2015-10-08 | 2015-12-23 | 广东欧珀移动通信有限公司 | Shooting method and device |
| CN105513067A (en) * | 2015-12-03 | 2016-04-20 | 小米科技有限责任公司 | Image definition detection method and device |
| CN105844641A (en) * | 2016-03-24 | 2016-08-10 | 武汉工程大学 | Adaptive threshold segmentation method in dynamic environment |
| CN106326824A (en) * | 2015-07-02 | 2017-01-11 | 阿里巴巴集团控股有限公司 | Method and device for automatically screening pictures |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102087652A (en) * | 2009-12-08 | 2011-06-08 | 百度在线网络技术（北京）有限公司 | Method for screening images and system thereof |
| CN103927749A (en) * | 2014-04-14 | 2014-07-16 | 深圳市华星光电技术有限公司 | Image processing method and device and automatic optical detector |
| CN104683780B (en) * | 2015-03-24 | 2017-01-04 | 广东本致科技有限公司 | The auto white balance method of a kind of video monitoring camera and device |
| CN105049728A (en) * | 2015-08-17 | 2015-11-11 | 广东欧珀移动通信有限公司 | Method and device for acquiring shot image |
| CN105761256B (en) * | 2016-02-05 | 2019-02-26 | 成都康烨科技有限公司 | The sub-pixel edge straight line acquisition methods and device of image |
-
2017
- 2017-06-20 CN CN201710467748.6A patent/CN107424128B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103501406A (en) * | 2013-09-16 | 2014-01-08 | 北京智谷睿拓技术服务有限公司 | Image collecting system and image collecting method |
| CN106326824A (en) * | 2015-07-02 | 2017-01-11 | 阿里巴巴集团控股有限公司 | Method and device for automatically screening pictures |
| CN105049875A (en) * | 2015-07-24 | 2015-11-11 | 上海上大海润信息***有限公司 | Accurate key frame extraction method based on mixed features and sudden change detection |
| CN105049734A (en) * | 2015-08-31 | 2015-11-11 | 广州市幸福网络技术有限公司 | License camera capable of giving shooting environment shooting prompt and shooting environment detection method |
| CN105187728A (en) * | 2015-10-08 | 2015-12-23 | 广东欧珀移动通信有限公司 | Shooting method and device |
| CN105513067A (en) * | 2015-12-03 | 2016-04-20 | 小米科技有限责任公司 | Image definition detection method and device |
| CN105844641A (en) * | 2016-03-24 | 2016-08-10 | 武汉工程大学 | Adaptive threshold segmentation method in dynamic environment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107424128A (en) | 2017-12-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA2730720C (en) | Apparatus and method for imaging the eye | |
| JP4865257B2 (en) | Fundus photographing apparatus and program | |
| US9521950B2 (en) | Apparatus and method for imaging an eye | |
| US20210353144A1 (en) | Wide Field Fundus Camera with Auto-Montage at A Single Alignment | |
| KR101998595B1 (en) | Method and Apparatus for jaundice diagnosis based on an image | |
| US20170161892A1 (en) | A method for acquiring and processing images of an ocular fundus by means of a portable electronic device | |
| EP3086705A1 (en) | Wide field retinal image capture system and method | |
| JP2015029865A (en) | Image processing apparatus and method, fundus image processing apparatus, image capturing method, fundus image capturing apparatus and method | |
| CN110916608B (en) | Diopter detection device | |
| CN106073698B (en) | A kind of fundus imaging method based on Android | |
| CN107424128B (en) | Method and system for extracting clear image frame from recorded eye fundus video | |
| JP2014161439A (en) | Eyeground information acquisition device and method, and program | |
| WO2015037316A1 (en) | Organ-imaging device and organ-imaging method | |
| JP2014233484A (en) | Ophthalmologic photographing apparatus and image forming method thereof | |
| CN113744254B (en) | Fundus image analysis method, fundus image analysis system, storage medium and computer equipment | |
| CN112669357B (en) | Fundus image synthesis method and fundus imager | |
| JP2003210409A (en) | Fundus camera | |
| JP5465210B2 (en) | Fundus photographing device | |
| TWI839124B (en) | Optical coherence tomography (oct) self-testing system, optical coherence tomography method, and eye disease monitoring system | |
| Giancardo et al. | Automatic quality enhancement and nerve fibre layer artefacts removal in retina fundus images by off axis imaging | |
| CN106961560A (en) | A kind of underwent eye-ground vascular fluorescence visualization dynamic video preparation method | |
| CN108784646A (en) | A kind of auxiliary focusing cylinder and focusing assisted Selection system, method | |
| KR20230081189A (en) | Pupillary light reflex measurement method using cell phone | |
| WO2017167587A1 (en) | Unobtrusive fundus imaging | |
| TW201808210A (en) | Endoscope imaging system and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |