WO2018117468A1

WO2018117468A1 - Image providing method and device for acquiring and storing intelligent high resolution endoscopic image

Info

Publication number: WO2018117468A1
Application number: PCT/KR2017/013898
Authority: WO
Inventors: 장민혜; 강동구; 진승오; 배영민; 신기영
Original assignee: 한국전기연구원
Priority date: 2016-12-20
Filing date: 2017-11-30
Publication date: 2018-06-28
Also published as: KR20180071532A

Abstract

The present invention relates to an image providing method and device for acquiring and storing an intelligent high resolution endoscopic image, the image providing device of the present invention including: an image processing device for processing an image; an image buffer for allowing the image processing device to store a still image of a prescribed unit of time from a moving image acquired by an image acquisition device; and a storage device for storing a still image corresponding to a still image capture request, wherein for one or more images among images stored in the image buffer or storage device, the image providing device assists a user so they are capable of determining whether or not to perform re-imaging by visually or acoustically providing alarm information about a determination result regarding the presence or absence of a blur, or the location of a blur.

Description

Image providing method and device for acquiring and storing intelligent high resolution endoscope images

The present invention relates to a method and apparatus for providing a still image capture during video recording, the function of determining the presence or absence of blur so that the user can re-photograph the desired image directly when the still image capture request, such as endoscope image capture An image providing method and apparatus capable of acquiring and storing an intelligent high resolution endoscope image capable of providing the same.

During video image capture, such as an endoscope image, the user may wish to capture still images. However, global or local blur may occur when capturing still images using a camera in low light conditions such as a dark room, when a part of the organ such as the gastrointestinal tract is moving when shooting an endoscope image, or due to camera shake during shooting. Symptoms may occur.

Various De-Blurring techniques have been developed to compensate for blurred images in general images.However, in medical images such as endoscopes, information that may cause errors in the diagnosis process may be added. The blurring technique cannot be applied to medical images.

In order to solve this problem, the conventional image processing apparatus continuously stores a still image in an image buffer during video image capturing, and automatically receives one of all still images stored before a certain time regardless of the user's intention when an image capture request is received. Most conventional image processing apparatuses recently developed have previously stored still images in a buffer for a predetermined time (Δt) as shown in FIG. 1, and when a still image capture request is received, the sharpest image among them is selected. To automatically select and save. In this case, when the motion blur occurs in the camera, the blur may occur in the still images stored before the capture request, and thus, even though the clearest image is selected among the pre-stored images, no meaningful image may be selected.

Therefore, the conventional image processing apparatus provides a captured image that is not directly selected by the user without reflecting the intention of the user, and thus, there is a problem in that a blurred image can be provided as the captured image.

As related prior documents, US Patent Publication No. US 2014/0132746 A1 (Image Capture Stabilization) and the like can be referred to.

Accordingly, the present invention has been made to solve the above-described problems, an object of the present invention, unlike the conventional method of automatically selecting a representative image regardless of the user's intention when the still image capture request during the recording of the endoscope image, etc. In addition, the present invention provides a method and apparatus for providing an image capable of assisting a user to check the blur and to re-determine whether necessary to retake the image if necessary.

First, to summarize the features of the present invention, an image providing apparatus according to an aspect of the present invention for achieving the above object, the image processing device for image processing; An image buffer for storing, by the image processing apparatus, a still image of a predetermined time unit among the moving images acquired by the image capturing apparatus; And a storage device for storing a still image corresponding to a still image capture request, and for the one or more images among the image buffer or the image stored in the storage device, the image processing apparatus includes a blur or a blur position. Alarm information on the determination result may be visually or audibly provided to assist the user in determining whether to retake the image.

In response to the still image capture request, the image processing apparatus selects one or more images determined to be less blurry among the one or more still images stored in the image buffer as an image to be provided as a captured still image. Can be stored on the storage device.

The image processing apparatus may alarm one or more images of one or more still images stored in the image buffer and one or more still images stored in the storage device as a still image corresponding to the still image capture request, with respect to the still image capture request. It can be stored in the storage device with or without.

The image processing apparatus stores one or more still images re-photographed through the image capturing apparatus in the storage device according to a re-shooting request, and uses one or more still images newly acquired through the image capturing apparatus and stored in the image buffer. By re-determining the presence or absence of the blur, the process of storing the re-photographed still image in the storage device or re-photographing may be repeatedly performed.

The image processing apparatus may detect a region of interest in a first method and an entire region of a first region, which is performed for each region divided into a predetermined block on a test target image to determine whether the blur is present or the position of the blur, and then apply the image to a corresponding region of interest. A second method performed for each divided region, a first method combining the first method and the second method, and performing a divided method for each divided region and giving more weight to the region of interest, or when the still image capture request occurs Any one or more of a fourth method of matching and comparing temporally adjacent images may be used by referring to the still image stored in the image buffer and one or more still images stored in the image buffer before the corresponding still image capture request is received. .

The image processing apparatus may provide a visual alarm by a method of differently displaying or highlighting an edge color of the image whether the still image to be provided is a normal image without blur or the still image to be provided is a blur image with blur. Alternatively, when the still image to be provided is a blur image, a visual alarm may be provided for the blur position by displaying the area of the blur image in the still image in a predetermined color together with the position of the still image of the entire video. Can be.

If the image processing apparatus continuously retakes images by blur detection, and the number of re-photographing requests or the number of still images stored in the storage device after the still image capturing request is more than a predetermined number, the alarm mode ends. After the still image capture request, a clean image having the least blur among the images stored in the storage device may be selected and stored in the storage device as an image to be provided as the captured still image.

If the image processing apparatus continuously retakes images by blur detection, and the number of re-photographing requests or the number of still images stored in the storage device after the still image capturing request is more than a predetermined number, the alarm mode ends. According to a user's request, a plurality of images to be provided as the captured still image stored in the storage device after the still image capturing request may be identified and provided as a group.

The image processing apparatus may provide an alarm indicating that the image includes medically beneficial information by performing a medical value determination of whether the corresponding image includes medically beneficial information with reference to a medical image database of the medical value determining apparatus. .

The medical value determination may be made after or before the determination of the presence or absence of the blur.

If the medical value determination is made prior to the determination of the presence or absence of the blur, an alarm that includes the medically beneficial information immediately after detection of the presence or absence of the blur or without detection of the presence or absence of the blur It can also provide

The medical value determination device may accumulate and maintain learned data about the medically beneficial information in the medical image database according to a predetermined learning algorithm.

The image providing method of providing a still image captured by the image processing apparatus during the capturing of the moving image includes blurring of one or more still images stored in the image buffer by the image processing apparatus in response to a still image capture request during the moving image capturing by the image capturing apparatus. Selecting one or more images determined to be low as an image to be provided as a capture still image and storing the captured image in a storage device; And assisting the user in determining whether to retake the image by providing visual or audio information about an image to be provided as the still image captured by the image processing apparatus, and alarm information regarding a result of the determination of the presence or absence of the blur. It includes.

According to the image providing method and apparatus according to the present invention, when providing a still image capture request during the shooting of the endoscope image, such as an alarm for blur or not, the user can check whether the blur, and if necessary re-shooting to determine whether to re-shoot itself By assisting to overcome the limitations of the prior art in which an image irrespective of a user's intention may be stored or a blur image without information may be stored, and only meaningful information may be efficiently generated and provided.

1 is a view for explaining a still image capture method in a conventional image processing apparatus.

2 is a block diagram of an image providing apparatus according to an exemplary embodiment.

3 is a diagram for describing a method of performing a blur test for each region divided according to the present invention.

4 is a view for explaining a blur test performing method based on the ROI detection in the present invention.

FIG. 5 is a diagram for describing a method of detecting blur by using a temporal collation method according to the present invention.

FIG. 6 is a view for explaining a relationship between acquisition speeds of temporal contrast target images stored in an image buffer with respect to a camera movement speed.

7 is an example for explaining a visual alarm method for a normal image and a blur image in the present invention.

8 is an example for explaining a visual alarm method of the blur position in the present invention.

9 is a view for explaining various alarm schemes provided to a user in the present invention.

Hereinafter, with reference to the accompanying drawings will be described in detail with respect to the present invention. In this case, the same components in each drawing are represented by the same reference numerals as much as possible. In addition, detailed descriptions of already known functions and / or configurations are omitted. The following description focuses on parts necessary for understanding the operation according to various embodiments, and descriptions of elements that may obscure the gist of the description are omitted. In addition, some components of the drawings may be exaggerated, omitted, or schematically illustrated. The size of each component does not entirely reflect the actual size, and thus the contents described herein are not limited by the relative size or spacing of the components drawn in the respective drawings.

2 is a block diagram of an image providing apparatus 100 according to an exemplary embodiment.

Referring to FIG. 2, the image providing apparatus 100 according to an exemplary embodiment of the present invention may include an image obtaining apparatus 110 such as a digital camera, an endoscope camera, an image processing apparatus 120 performing an overall function of image processing, The image buffer 130 stores a still image (data) stored in the storage device 140 by a still image or a previous still image capture request for a predetermined time, and a storage device 140 such as a memory. The image processing apparatus 120, the image buffer 130, and the storage device 140 may be implemented in hardware such as a semiconductor processor, software such as an application program, or a combination thereof.

In the present invention, the image capturing apparatus 110 acquires a moving image (video) (data), the image processing apparatus 120 may control to display the moving image on the display device, and the still image (data) by a predetermined time unit ) Is stored in the image buffer 130. The image processing apparatus 120 may store and maintain a still image for a predetermined time in the image buffer 130 and delete an image exceeding a predetermined capacity as time passes.

If a still image capture request is generated from a user or the like while the image capturing apparatus 110 is capturing a video, the image processing apparatus 120 may convert the image buffer 130 from the one or more still images stored in the image buffer 130 to the image buffer 130. A clean image having the least blur among one or more stored still images is selected and stored in the storage device 140, and provided through the display device according to a user's request. Or, as described below, when a still image capture request occurs, one or more still images stored in the image buffer 130 and one or more still images stored in the storage device 140 as still images corresponding to the still image capture request. One or more images may be stored in the storage device with or without an alarm.

In this case, the image processing apparatus 120 may select the clear image having the least blur with reference to the information acquired by the acceleration sensor and the Gyro sensor. In this case, the image processing apparatus 120 may determine whether the blur is present or the blur position, and as a result of determining whether the image is a blur image or a normal image, output information, such as a blur position, as visual alarm information through a display device, a speaker, or the like. It outputs audible alarm information to assist the user in deciding whether to retake. The user may input an image capture request to the image processing apparatus 120 through a foot switch, a button, a speech, a joystick, a bio signal such as brain waves or safety.

However, the user may check the visual or audio alarm information as described above, and if the user is not satisfied due to blur, may decide to retake and request a reshooting, and the re-shot still image is stored according to the re-shooting request of the user. To be stored in the device 140. That is, the image processing apparatus 120 may be obtained from one or more still images (possibly utilizing still images stored previously) which are acquired by the image capturing apparatus 110 and newly stored in the image buffer 130 according to a user's re-shooting request. By re-determining the presence or absence of the blur, the process of storing the re-photographed still image in the storage device 140 or re-taking again may be repeatedly performed.

On the other hand, the image processing apparatus 120 performs a blur test on the target still image for determining the presence or absence of blur or for each region divided into predetermined blocks (for example, 128 * 128 pixels, etc.) for the target still image. It may be determined whether the image is a blur image or a normal image. The blur test uses a hand-crafted feature, such as the difference in gradation value with neighboring pixels in each pixel, or a dip that calculates the convolution value between the patch image and the kernels. This can be done using the deep learning method. Related prior studies include "Blur detection for digital images using wavelet transform", Hanghang Tong, Mingjing Li, Hongjinang Zhang, Changshui Zhang, 2004 IEEE International Conference on Multimedia and Expo (ICME), June 2004, "Convolutional Neural Networks for No-Reference. Image Quality Assessment ", 2014 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June 2014, and the like. In addition, the image processing apparatus 120 may perform a blur test with reference to signals acquired from the acceleration sensor and the gyro sensor included in the image capturing apparatus.

3 is a diagram for describing a method of performing a blur test for each region divided according to the present invention. For example, when the blur test is performed for each region divided into a predetermined block (for example, 128 * 128 pixels, etc.) with respect to the test target image, as shown in FIG. The degree of blur may be calculated based on the difference to determine whether the blur image is determined according to the number of blocks exceeding the reference value.

On the other hand, a blur test for each divided region may be performed to distinguish between global movement caused by camera movement and local movement caused by some long-term movement.

FIG. 4 is a diagram for describing a blur test performing method based on detection of a region of interest in the present invention. The region of interest may include a region that shows medical abnormality. The ROI may be detected by the image processing apparatus 120 using a method using a hand-crafted feature, a method using deep learning, or a user's eye tracking information. For example, the image processing apparatus 120 may first identify a method of finding a portion having a difference in gray level change degree on the left and right of a symmetrical object as shown in FIG. Method, or a deep neural network trained with a large number of images containing the region of interest, and then detects the region of interest that contains the region and then blurs the blur region above the region of interest. The method may determine whether a blur image is applied.

Alternatively, the region of interest is first detected as shown in FIG. 4, and then a blur test is performed for each region divided into predetermined blocks (eg, 128 * 128 pixels) as shown in FIG. 3. The block may be given more weight to determine whether a blur image is present.

FIG. 5 is a diagram for describing a method of detecting blur by using a temporal matching method in the present invention.

For example, the image processing apparatus 120 may detect blur by using a temporal matching method with reference to a plurality of images stored in the image buffer 130 as shown in FIG. 5. In other words, the coordinates of one or more objects included in a plurality of temporally adjacent images stored in the image buffer 130 are matched, and the sharpness of the corresponding portion is compared by comparing the gray level value or the degree of change thereof. It may be determined whether the blur image. In this case, the temporal contrast target images stored in the image buffer 130 may include a still image stored in the image buffer 130 when the still image capture request occurs, and one or more still images stored in the image buffer 130 before the corresponding still image capture request is received. The image may include one or more still images stored in the storage device 140 by a previous still image capture request.

The still images are stored in the image buffer 130 at predetermined time units. As illustrated in FIG. 6, the acquisition speed of the still image transferred from the image acquisition device 110 and stored in the image buffer 130 may vary depending on the speed of the camera 110 movement. For example, when the camera 110 moves at a high speed, only a small number of similar still images are stored in the image buffer 130 at the same time, and when the camera moves at a slow speed, a large number of similar still images are recorded at the same time as the still image. Transfer it to the buffer 130 and store it. The image processing apparatus 120 may store and maintain a still image for a predetermined time in the image buffer 130 and delete an image exceeding a predetermined capacity as time passes.

7 is an example for explaining a visual alarm method for a normal image and a blur image in the present invention. 8 is an example for explaining a visual alarm method of the blur position in the present invention.

When the blur is detected in various ways as described above, the image processing apparatus 120 provides the blur detection result such as the presence or absence of the blur, the blur position through a visual or audio alarm, and the visual alarm color as shown in FIG. 7. After the blurring or position detection as described above through image information extraction (see S120 of FIG. 9) for the still image (see S110 of FIG. 9) to be provided, using the method of displaying differently (See S130 of FIG. 9), the user may be informed whether the normal image without blur (eg, green), or the still image to be provided is a blur image with blur (eg, red) (FIG. 9, see S140). In addition, as shown in FIG. 8, when the still image to be provided for the corresponding still image capture request is a blur image, the still image together with the position (eg, time information, etc.) of the still image among the entire video images (movies) is included in the still image. The area of the location where the blur is detected may be displayed in a predetermined color (for example, red color) to provide a visual alarm on the blur location. Instead of the same color, the same highlighting (eg, displaying a blur image brighter than a still image) may also indicate whether a blur image is present.

The user may check the visual or auditory alarm information as described above, and if the user is not satisfied due to the blur, may determine the re-shooting and request the re-shooting, and may cause the re-photographed still image to be stored in the storage device 140. . That is, in response to a re-photographing request of a user or the like, the image processing apparatus 120 is obtained from one or more still images (possibly using still images stored previously) that are acquired by the image capturing apparatus 110 and newly stored in the image buffer 130. The process of re-determining the presence or absence of the blur and storing or re-taking the corresponding still image may be performed.

In some cases, the image processing apparatus 120 stores the database information stored by dividing the blurred image and the non-blurred image in the storage device 140 or a separate memory, in order to detect the blur. As described above, by analyzing the feature points and gray values of the images determined as the newly updated blur image and the images determined as the normal images, machine learning (eg, Deep Learning) is performed to further increase the user's intention. It is reflected to improve the detection accuracy of the blur image.

However, when the blur detection is continuously performed even after re-taking several times due to the continuous movement of the subject or the subject, such as when taking an endoscope image of the patient, the storage device after the number of re-photographing requests or after the still image capturing request If the number of still images (frames) stored in the 140 is greater than or equal to a predetermined number, the alarm mode for determining the blur detection (blur or blur position) is terminated as described above, and the storage device 140 is requested after the still image capture request. The clearest image having the least blur among the stored images may be selected and stored as the image to be provided as the captured still image in the storage device 140 and provided to the display device. Alternatively, when the alarm mode is terminated as described above, when a plurality of still images are stored as images to be provided as a capture still image to the storage device 140 after a request for capturing still images by a continuous alarm is detected and the continuous alarm is detected. In response to the request, the console may identify the group as one group so that the entire image may be provided through the display device.

Meanwhile, the conventional image processing apparatus arbitrarily selects and stores the clearest image without reflecting the intention of the user. However, this is simply selecting the sharpest image from the stored images, and there is no guarantee that the selected image is clear enough to be suitable for medical diagnosis purposes. In addition, even if a blur exists in a part of the acquired image, if the part without the blur is a part necessary for diagnosis, the image may be of medical value, so that the image may be stored or the user may determine whether to store it. It is necessary to tell the presence and location.

Therefore, for this case, for example, the image processing apparatus 120 may extract the medical information through the image information extraction (see S120 of FIG. 9) for the still image (see S110 of FIG. 9) as described above. With reference to the medical image database of the device (hereinafter medical value determination device) for determining whether or not to include the beneficial information to determine whether the image contains medically beneficial information (see S150 in FIG. 9), When medically beneficial information (eg, polyp position, etc.) is included, an alarm function may be provided to inform the user that the medically beneficial information is included (see S160 of FIG. 9). Determination of whether the medical information includes the above information (hereinafter medical value judgment), the presence or absence of blur or after detecting the position (see S130 of FIG. 9) when the still image to be provided for the still image capture request is a blur image The medical value may be determined immediately after extracting the image information (see S120 of FIG. 9). In the case where the medical value judgment is performed first, an alarm function may be provided after the blurring or the position detection process.

As described above, the method of providing an alarm to the user by the image processing apparatus 120 may be summarized as follows. As shown in FIG. 9, first, after extracting image information from an input image, the presence or absence of a blur is detected from the extracted feature information, and an alarm may be provided to the user. Alternatively, the medical value may be directly determined using the image information extracted from the input image, and then an alarm may be provided depending on whether the medical value is determined. Alternatively, the presence or absence of a blur may be detected using image information extracted from the input image, and then the medical value may be determined using this information to provide an alarm. Alternatively, the medical value may be determined using image information extracted from the input image image, and then an alarm may be provided after detecting the presence or absence of blur at the medical value.

The image processing apparatus 120 may determine whether there is blur with respect to the obtained image by using a general image quality evaluation algorithm or a blur detection algorithm, and may grasp the location of the blur. In the case of medical images, it is difficult to obtain a clear image without blur at the same location, and thus a no-reference image quality assessment algorithm can be applied. This algorithm uses transforms such as Discrete Wavelet Transform, Discrete Cosine Transform, or extracts features of color, contrast, and edge information, and then machine learning such as Support Vector Machine, Support Vector Regression, Random Forest, and Neural Network. Techniques can be used to determine the presence, extent, and location of blur.

On the other hand, whether or not the image contains medically beneficial information is subject to the subjective evaluation of the medical staff, and thus the medical value of the image cannot be determined using a general blur detection and image quality evaluation algorithm or a general image database. Therefore, the medical value judgment device should use a medical image database that displays information that the medical staff has directly annotated whether or not it contains medically beneficial information, and the medical value judgment is made using the medical image and the display result of the medical image database. You can train the device.

A medical imaging database indicating whether a medical staff includes medically beneficial information is in the form of / without medically beneficial information in addition to the medical image, the location of the medically beneficial information (eg, the location of the polyp), or its Scores for degree, and the like.

The medical value database may be configured by various methods used in general blur detection and image quality evaluation algorithms using the medical image database of the medical value determination device. For example, you can use the Discrete Wavelet Transform, Discrete Cosine Transform, or other features to extract color, contrast, or edge information, and then use the benefit information, degree, or location to support Vector Machine, Support Vector Regression. It is possible to accumulate the learned data in the medical image database according to a predetermined learning algorithm such as, Random Forest, Neural Network. For example, when the blur position is detected or compared with the position where the medical image database has the beneficial information, the alarm may be provided to the user when the medical image database includes the position having the medical information.

Such a function used for input / output data processing in the image processing apparatus 120 according to an embodiment of the present invention may be implemented as computer readable codes on a recording medium readable by a device such as a computer. A combination of a recording medium and a device such as a computer can be implemented to input, output, and display data or information necessary to perform a function. The computer-readable recording medium includes all kinds of recording devices in which data that can be read by a computer system is stored. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage, hard disk, removable storage, and also carrier wave (for example, transmission over the Internet). It also includes the implementation in the form of. In addition, the computer-readable recording medium may be distributed in a computer system connected through a network (eg, the Internet, a mobile communication network, etc.), and may have a form in which computer-readable codes are stored in a distributed manner and may be executed through a network.

As described above, according to the image providing apparatus 100 according to the present invention, by providing an alarm for the blur when a still image capture request during the recording of the endoscope image, the user can check whether the blur and if necessary to retake By assisting the user to re-determine and retake, he overcomes the limitations of the prior art that can store images irrelevant to the user's intention or blur images without information, and can efficiently create and provide only meaningful information. have.

In the present invention as described above has been described by the specific embodiments, such as specific components and limited embodiments and drawings, but this is provided to help a more general understanding of the present invention, the present invention is not limited to the above embodiments. For those skilled in the art, various modifications and variations may be made without departing from the essential features of the present invention. Accordingly, the spirit of the present invention should not be limited to the described embodiments, and all technical ideas having equivalent or equivalent modifications to the claims as well as the following claims are included in the scope of the present invention. Should be interpreted as.

Claims

An image processing apparatus for image processing;

An image buffer for storing, by the image processing apparatus, a still image of a predetermined time unit among the moving images acquired by the image capturing apparatus; And

A storage device for storing a still image corresponding to the still image capture request;

For one or more images among the images stored in the image buffer or the storage device, the image processing apparatus may visually or auditoryly provide alarm information on a result of the determination of the presence or absence of blur and whether the user retakes the image. The image providing device, characterized in that assists to determine.
The method of claim 1,

The image processing apparatus,

In response to the still image capturing request, the image processing apparatus selects one or more images determined to be less blurry among the one or more still images stored in the image buffer as an image to be provided as a captured still image, and stores them in the storage device. An image providing device characterized in that.
The method of claim 1,

The image processing apparatus may alarm one or more images of one or more still images stored in the image buffer and one or more still images stored in the storage device as a still image corresponding to the still image capture request, with respect to the still image capture request. The image providing device, characterized in that the storage in the storage device regardless of the presence or absence.
The method of claim 1,

The image processing apparatus stores the still image re-photographed through the image obtaining apparatus in the storage device in response to the re-shooting request.

Re-determining the presence or absence of the blur or the blur position by using one or more still images newly acquired by the image acquisition device and stored in the image buffer to store or retake the re-photographed still image in the storage device. Image providing apparatus, characterized in that for repeating.
The method of claim 1,

The image processing apparatus, with respect to the test target image in order to determine whether the blur or blur position,

A first method performed for each area divided into predetermined blocks,

A second method of first detecting a region of interest in the entire region and then performing the divided regions for the region of interest;

A third method combining the first method and the second method and performing each segmented area to give more weight to the ROI; or

A fourth method of matching and comparing temporally adjacent images by referring to a still image stored in the image buffer when the still image capture request occurs and at least one still image stored in the image buffer before the corresponding still image capture request is received;

Image providing apparatus, characterized in that using any one or more.
The method of claim 1,

The image processing apparatus,

Provide a visual alarm by displaying or highlighting the edge color of the image differently whether the still image to be provided is a normal image without blur, or whether the still image to be provided is a blur image with blur,

When the still image to be provided is a blur image, a visual alarm is provided for the blur position by displaying a region of a blur detected position in the still image in a predetermined color together with the position of the still image of the entire video. Image providing apparatus.
The method of claim 1,

The image processing apparatus,

When the reshooting is continuously performed by blur detection, and the number of re-shooting requests or the number of the still images stored in the storage device after the still image capturing request is more than a predetermined number, the alarm mode is terminated and the still image capturing request is performed. And a clean image having the least blur among the images stored in the storage device is selected and stored in the storage device as an image to be provided as the captured still image.
The method of claim 7, wherein

The image processing apparatus,

When re-shooting is performed continuously by blur detection, and the number of re-photographing requests or the number of still images stored in the storage device after the still image capturing request is more than a predetermined number, the alarm mode is terminated, and And a plurality of images to be provided as the captured still image stored in the storage device as one group after the still image capture request is provided.
The method of claim 1,

The image processing apparatus may refer to the medical image database of the medical value determination device to perform a medical value determination of whether the corresponding image includes medically beneficial information, and provide an alarm indicating that the image includes medically beneficial information. Image providing apparatus.
The method of claim 9,

The medical value determination is performed after or before the determination of the presence or absence of the blur position.
The method of claim 10,

If the medical value determination is made prior to the determination of the presence or absence of the blur, an alarm that includes the medically beneficial information immediately after detection of the presence or absence of the blur or without detection of the presence or absence of the blur Image providing apparatus, characterized in that for providing.
The method of claim 9,

And the medical value determining apparatus accumulates and maintains the learned data on the medically beneficial information in the medical image database according to a predetermined learning algorithm.
In the image providing method for providing a still image captured by the image processing apparatus during video recording,

Responding to a still image capture request during video recording by the image capturing apparatus, the image processing apparatus selects one or more images determined to be less blurry among the one or more still images stored in the image buffer as the image to be provided as the captured still image, and the storage device. Storing in; And

Assisting the user in determining whether to retake the image by providing visual or audio information on the image to be provided as the captured still image, alarm information regarding a result of the blur or the blur position

Image providing method comprising a.