A smart medical interaction method and system based on big data

Abstract

The present invention relates to the field of data processing, and in particular to a smart medical interaction method and system based on big data. The method includes the following steps: obtaining a motion image, a background image, and a trajectory sequence in each channel image; obtaining a motion timing sequence in the motion image, and dividing the motion timing sequence into motion data segments; obtaining the deviation value of each pixel value in the motion data segment, and adjusting the length of the binary data of the deviation value of each pixel value in the motion data segment; obtaining the binary data of the deviation value of each pixel value in the background image and adjusting the length of the binary data; using the trajectory sequence, the adjusted data of the binary data of the deviation value of each pixel in the motion image block, and the adjusted data of the binary data of the deviation value of each pixel value in the background image as transmission objects to achieve compressed transmission. This reduces the amount of transmission of unnecessary data.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a smart medical interaction method and system based on big data. Background Art

[0002] In order to solve the problem that long distances hinder patients from seeking medical treatment, many medical institutions have introduced remote consultation technology. Patients describe their problems by shooting videos and other means, and doctors give diagnosis conclusions based on the video information, thus solving the problem that long-distance patients cannot seek medical treatment. In order to reduce the pressure of data transmission during remote consultation, data needs to be compressed.

[0003] When transmitting the video taken during the remote consultation, what is generally transmitted is a binary data sequence. Since the pixel values ​​of different pixels are different, the lengths of the corresponding binary data are different. If a separator is not inserted between the binary data of the pixel values ​​of different pixels, the binary data of the pixel values ​​of all pixels are spliced ​​together for transmission, and it is impossible to identify which position is the boundary position of the binary data of the pixel values ​​of different pixels, resulting in the inability to accurately read the pixel values ​​of each pixel; in order to solve this problem, the binary data of the pixel value of each pixel is generally adjusted to the maximum length required by the pixel value. This method will undoubtedly increase unnecessary transmission pressure. Therefore, how to reduce the transmission of useless data while identifying the boundary position of different pixel values ​​has become the research focus of the present invention.

[0004] The patent document with announcement number CN100380384C discloses a medical system architecture for interactive transmission and progressive display of compressed image data. The patent document focuses on group compression storage and convenient access. The method in the patent document cannot solve the technical problem of the present invention. Summary of the invention

[0005] In order to solve the problem of how to reduce the transmission of useless data while identifying the boundary positions of different pixel values, the present invention provides a smart medical interaction method and system based on big data.

[0006] In the first aspect, the present invention provides a smart medical interaction method based on big data, which adopts the following technical solutions:

[0007] A smart medical interaction method based on big data, comprising the steps of:

[0008] Obtain the smart medical video and each channel image of each frame image in the smart medical video;

[0009] Acquire the area where the moving object is located and the area where the non-moving object is located in the channel image obtained in advance, and record them as the moving image and the background image respectively; record the sequence constructed by the central coordinates of the moving image in all frame images as the trajectory sequence;

[0010] The sequence composed of pixel values ​​at a pixel position in the running images of all frame images is recorded as a motion timing sequence, and the motion timing sequence is divided into a number of data segments recorded as motion data segments; the difference between each pixel value in the motion data segment and the minimum value is recorded as a deviation value, and the length of the longest data obtained in the binary data of the deviation values ​​of all pixel values ​​in the motion data segment is recorded as a reference length; the binary data of the deviation value of each pixel value in the motion data segment is adjusted to the reference length by zero padding; the binary data of the deviation value of each pixel value in the background image is obtained, and the binary data of the deviation value of each pixel value in the background image is adjusted;

[0011] The trajectory sequence, the adjusted data of the binary data of the deviation value of each pixel in the motion image block, and the adjusted data of the binary data of the deviation value of each pixel value in the background image are taken as transmission objects to achieve compressed transmission.

[0012] The present invention performs compression analysis on the motion area and the background area in the channel image separately, and fully utilizes the characteristics of the similarity of the pixels at the same position in the motion area and the similarity of the pixels at the same position in the background area in the frame sequence, so that the binary data lengths of the pixel values ​​in the compressed and analyzed pixel value sequence are similar, providing a basis for effectively reducing the amount of unnecessary data transmission in the future; further, by performing segmented analysis on the pixel values ​​in the compressed and analyzed pixel value sequence, similar data are grouped together, and the similarity characteristics of the data are further improved; further, the deviation values ​​of the pixel values ​​with greater similarity are taken as compression processing objects, and the characteristics of the similar data with smaller deviation values ​​and smaller deviation value differences are fully utilized, so that the binary data of the compression processing objects are not only smaller in length, but also smaller in difference, providing a basis for reducing the amount of unnecessary data transmission in the future; further, by adjusting the length of the binary data of the deviation values ​​of the pixel values ​​with greater similarity, the difference between the adjusted length and the length before the adjustment is smaller, thereby ensuring that the boundary position between the deviation values ​​can be identified while reducing the amount of unnecessary data transmission.

[0013] Preferably, the step of obtaining the pre-obtained region where the moving object is located and the region where the non-moving object is located in the channel image, which are recorded as the moving image and the background image respectively, includes:

[0014] Perform grayscale processing on each frame image to obtain a grayscale image; use a frame difference method to process the grayscale images of all frame images to obtain a frame difference image of each frame image; obtain the bounding rectangle of the area composed of pixels whose grayscale values ​​are not equal to zero in the frame difference image; obtain the size corresponding to the largest bounding rectangle among the bounding rectangles in the frame difference images of all frame images and record it as the reference size;

[0015] In the channel image, an image area enclosed by a rectangle with a reference size and a center of a circumscribed rectangle in a frame difference image of a corresponding frame as the center is obtained and recorded as a motion image; and an area other than the motion image in the channel image is recorded as a background image.

[0016] The present invention obtains the moving area image by using the frame difference method, which is simple to operate and has high implementation efficiency.

[0017] Preferably, recording a sequence constructed by center coordinates of motion images in all frame images as a trajectory sequence includes:

[0018] In the smart medical video, a frame segment from the beginning of the appearance to the disappearance of a moving object is obtained, the center coordinates of all moving images corresponding to the moving object are obtained in the frame segment, and the sequence formed by them is recorded as a trajectory sequence.

[0019] The present invention locates the motion trajectory through the central coordinates, uses less data to characterize the position information of each frame of motion image, and provides a basis for reducing unnecessary data transmission volume.

[0020] Preferably, the method for acquiring the frame segment includes:

[0021] Acquire an individual region; acquire a reference individual region in a frame difference image of a next frame, and calculate the morphological similarity between the individual region and the reference individual region; multiply the morphological similarity by the inverse of the distance between the individual region and the reference individual region, and then perform normalization processing to obtain the probability of being the same individual; if the probability of being the same individual is greater than a preset probability threshold, it is determined that the moving object corresponding to the individual region and the reference individual region is the same moving object; if the probability of being the same individual is not greater than the preset probability threshold, it is determined that the moving object corresponding to the individual region and the reference individual region is not the same moving object; and acquire the frame number interval corresponding to the image in which the same moving object exists as the frame segment in which the same moving object begins to appear and disappears.

[0022] The present invention uses simple information such as shape and distance information to analyze whether they belong to the same moving object, and its operation is simpler and the implementation efficiency is higher.

[0023] Preferably, recording a sequence of pixel values ​​at a pixel position in the running images of all frame images as a motion time series sequence, and dividing the motion time series sequence into a plurality of data segments as motion data segments, comprises:

[0024] Obtain a frame segment from the beginning of the appearance to the disappearance of a moving object in the smart medical video, and record a sequence consisting of all pixel values ​​of a pixel position of all moving images in the frame segment as a motion time sequence;

[0025] The pixel values ​​in the motion time series are clustered to obtain several categories, and the pixels belonging to the same category in the motion time series are segmented into one data segment to obtain several data segments recorded as motion data segments.

[0026] Preferably, the step of adjusting the binary data of the deviation value of each pixel value in the motion data segment to a reference length by zero padding includes:

[0027] The length of the binary data of the deviation value of the pixel value in the motion data segment is compared with the reference length. If the length of the binary data of the deviation value of the pixel value in the motion data segment is less than the reference length, the length of the binary data of the deviation value of the pixel value in the motion data segment is adjusted to the reference length by adding several zeros before the binary data of the deviation value of the pixel value in the motion data segment. If the length of the binary data of the deviation value of the pixel value in the motion data segment is not less than the reference length, no zero padding is required.

[0028] The present invention adjusts the length by padding zeros in the front, so that the data size is not changed and the boundary positions of different data can be identified at the same time.

[0029] Preferably, the obtaining of binary data of the deviation value of each pixel value in the background image and adjusting the binary data of the deviation value of each pixel value in the background image comprises:

[0030] The sequence of pixel values ​​at a pixel position in the background image of all frame images is recorded as a background timing sequence, the background timing sequence is divided into several data segments recorded as background data segments, the binary data of the deviation value of each pixel value in the background data segment and the corresponding reference length are obtained, and the binary data of the deviation value of each pixel value in the background data segment is adjusted to the corresponding reference length.

[0031] Preferably, the adjusting data of binary data of deviation value of each pixel in the trajectory sequence, the motion image block and the adjusting data of binary data of deviation value of each pixel in the background image are used as transmission objects to realize compressed transmission, including:

[0032] The trajectory sequence, the adjusted data of the binary data of the deviation value of each pixel in the motion image block, and the adjusted data of the binary data of the deviation value of each pixel value in the background image are spliced ​​to obtain a transmission data sequence, and the transmission data sequence is used as a transmission object to achieve compressed transmission.

[0033] Preferably, the method for acquiring the transmission data sequence includes:

[0034] The sequence obtained by splicing together the binary adjusted data of the deviation values ​​of all pixel values ​​of each motion data segment is recorded as a unit sequence, and the minimum value of all pixel values ​​of the motion data segment, the number of pixel values ​​and the preset motion segmentation mark value are inserted before each unit sequence, and then the splicing process is performed to obtain a first sequence of a pixel position in the motion image; the preset motion pixel mark value is inserted before each first sequence, and then the splicing process is performed to obtain a transmission sequence of the motion image; the transmission sequence of the background image is obtained; the starting frame number and the ending frame number corresponding to the trajectory sequence are inserted before the trajectory sequence, and then the reference size, the transmission sequence of the motion image and the transmission sequence of the background image are spliced ​​together as the transmission sequence of the channel image.

[0035] In a second aspect, the present invention provides a smart medical interaction system based on big data, which adopts the following technical solutions:

[0036] A smart medical interaction system based on big data includes: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned smart medical interaction method based on big data is implemented.

[0037] By adopting the above technical solution, the above-mentioned smart medical interaction method based on big data is generated into a computer program and stored in a memory to be loaded and executed by a processor, so that a terminal device is made according to the memory and the processor for easy use.

[0038] The present invention has the following technical effects:

[0039] The present invention performs compression analysis separately on the motion area and the background area in the channel image, making full use of the similarity of pixels at the same position in the motion area and the similarity of pixels at the same position in the background area in the frame sequence, so that the binary data lengths of the pixel values ​​in the compressed and analyzed pixel value sequence are similar, providing a basis for effectively reducing the amount of unnecessary data transmission subsequently.

[0040] Furthermore, by performing segmented analysis on the pixel values ​​in the compressed pixel value sequence, similar data are grouped together, thereby further improving the similarity characteristics of the data.

[0041] Furthermore, the deviation values ​​of pixel values ​​with greater similarity are taken as compression processing objects, fully utilizing the characteristics of similar data with smaller deviation values ​​and smaller deviation value differences, so that the binary data of the compression processing object is not only smaller in length but also smaller in difference, thus providing a basis for subsequent reduction in unnecessary data transmission.

[0042] Furthermore, by adjusting the length of the binary data of the deviation values ​​of pixel values ​​with greater similarity, the difference between the adjusted length and the length before adjustment is smaller, thereby ensuring that the boundary position between the deviation values ​​can be identified while reducing unnecessary data transmission. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-restrictive manner, and the same or corresponding reference numerals represent the same or corresponding parts.

[0044] Figure 1 It is a flowchart of a method in a smart medical interaction method based on big data in an embodiment of the present invention. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0046] It should be understood that when the terms "first", "second", etc. are used in the claims, descriptions, and drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections.

[0047] The embodiment of the present invention discloses a smart medical interaction method based on big data, referring to Figure 1 , comprising steps S1 to S5:

[0048] S1: Acquire the smart medical video and each channel image of each frame image in the smart medical video.

[0049] Specifically, the video collected when the patient is having an online consultation is recorded as a smart medical video, and each channel image of each frame image in the smart medical video is obtained.

[0050] S2: Acquire the area where the moving object and the area where the non-moving object are located in the channel image obtained in advance, and record them as the moving image and the background image respectively; record the sequence constructed by the central coordinates of the moving image in all frame images as the trajectory sequence.

[0051] S20: Acquire the area where the moving object is located and the area where the non-moving object is located in the channel image obtained in advance, and record them as the moving image and the background image respectively.

[0052] It should be noted that since the color difference of the pixels of the moving object in the video is small in the frame sequence, the color difference of the pixels in other areas of the video except the area where the moving object is located is also small in the frame sequence. The smaller the color difference, the smaller the length of the corresponding binary data. If the binary data of the color difference is transmitted, it will reduce the large data transmission volume. Therefore, the color difference of the moving object in the frame sequence and the color difference of other areas in the frame sequence except the area where the moving object is located can be used as the transmission object to achieve data compression. First, the moving object and other areas must be separated.

[0053] Preferably, as an example, obtaining the area where the moving object is located and the area where the non-moving object is located in the channel image obtained in advance, and recording them as the moving image and the background image respectively, includes:

[0054] Perform grayscale processing on each frame image to obtain a grayscale image; use a frame difference method to process the grayscale images of all frame images to obtain a frame difference image of each frame image; obtain the bounding rectangle of the area composed of pixels whose grayscale values ​​are not equal to zero in the frame difference image; obtain the size corresponding to the largest bounding rectangle among the bounding rectangles in the frame difference images of all frame images and record it as the reference size;

[0055] In the channel image, an image area enclosed by a rectangle with a reference size and a center of a circumscribed rectangle in the frame difference image of the corresponding frame as the center is obtained and recorded as a motion image; and an area other than the motion image in the channel image is recorded as a background image.

[0056] S21: The sequence constructed by the center coordinates of the motion images in all frame images is recorded as a trajectory sequence.

[0057] It should be noted that, since the position of the motion image is different in each frame, in order to ensure that the pixels of the motion image area of ​​each frame can be restored during decompression, the movement trajectory of the motion image in the frame sequence needs to be recorded.

[0058] Preferably, as an example, a sequence constructed by the center coordinates of the motion images in all frame images is recorded as a trajectory sequence, including:

[0059] In the smart medical video, a frame segment from the beginning of the appearance to the disappearance of a moving object is obtained, the center coordinates of all moving images corresponding to the moving object are obtained in the frame segment, and the sequence formed by them is recorded as a trajectory sequence.

[0060] It should be added that in the smart medical video, the frame segment from the beginning of the appearance to the disappearance of a moving object is obtained, including:

[0061] In a frame difference image of one frame, a closed area composed of non-zero pixels is obtained and recorded as an individual area. For the tangent direction of each non-zero pixel in the individual area, a sequence composed of the occurrence frequencies of all kinds of tangent directions in the individual area is recorded as a morphological sequence. In a frame difference image of the next frame, an individual area closest to the individual area of ​​the frame difference image of the frame is obtained and recorded as a reference individual area. The cosine similarity between the morphological sequence corresponding to the individual area and the morphological sequence corresponding to the reference individual area is calculated and recorded as morphological similarity.

[0062] After multiplying the morphological similarity by the inverse of the distance between the individual region and the reference individual region, normalization is performed, and the normalized data is recorded as the probability of the same individual. If the probability of the same individual is greater than the preset probability threshold, it is determined that the moving object corresponding to the individual region and the reference region is the same moving object; if the probability of the same individual is not greater than the preset probability threshold, it is determined that the moving object corresponding to the individual region and the reference region is not the same moving object. This embodiment is described by taking the preset probability threshold of 0.8 as an example, and other embodiments may take other values, which are not specifically limited in this embodiment.

[0063] The frame number interval corresponding to the image containing the same moving object is obtained as the frame segment from the beginning of appearance to the disappearance of the same moving object.

[0064] S3: Record the sequence composed of pixel values ​​of a pixel position in the running images of all frame images as a motion timing sequence, divide the motion timing sequence into several data segments recorded as motion data segments; record the difference between each pixel value in the motion data segment and the minimum value as a deviation value, obtain the longest data length in the binary data of the deviation values ​​of all pixel values ​​in the motion data segment as a reference length; adjust the binary data of the deviation value of each pixel value in the motion data segment to the reference length by padding with zeros; obtain the binary data of the deviation value of each pixel value in the background image, and adjust the binary data of the deviation value of each pixel value in the background image.

[0065] S30: Record a sequence consisting of pixel values ​​at a pixel position in the running images of all frame images as a motion time series sequence, and divide the motion time series sequence into a plurality of data segments recorded as motion data segments.

[0066] It should be noted that in order to further reduce the transmission amount, it is necessary to ensure that the difference in pixel value in the frame sequence of the motion image is as small as possible. Therefore, pixels with similar values ​​can be segmented together, and then the pixel value difference is calculated in each segment and transmitted.

[0067] Preferably, as an example, a sequence of pixel values ​​at a pixel position in the running images of all frame images is recorded as a motion time series sequence, and the motion time series sequence is divided into a plurality of data segments recorded as motion data segments, including:

[0068] In the smart medical video, a frame segment from the beginning of the appearance to the disappearance of a moving object is obtained, and a sequence consisting of pixel values ​​at a pixel position in the running images of all frame images in the frame segment is recorded as a motion time series sequence; the DBSCAN clustering method is used to cluster the pixel values ​​in the motion time series sequence to obtain several categories, and the pixels belonging to the same category in the motion time series sequence are divided into a data segment to obtain several data segments recorded as motion data segments.

[0069] It can be seen that the motion data sequence contains the color values ​​of the same moving object in time series, so the data difference in the motion data sequence is small. By further segmenting the data with small differences, the data difference in the motion data segment is made smaller. This link provides the basis for subsequent data compression to a greater extent.

[0070] S31: obtaining the longest data length in the binary data of the deviation values ​​of all pixel values ​​in the motion data segment and recording it as a reference length; adjusting the binary data of the deviation value of each pixel value in the motion data segment to the reference length by padding with zeros.

[0071] It should be noted that, due to the differences between different pixel values ​​in the same motion data segment, the lengths of the binary data of the difference values ​​are also different, and when binary data of different lengths are spliced ​​together, the boundary position between the binary data cannot be identified. Therefore, it is necessary to adjust each difference binary data to the same length.

[0072] Preferably, as an example, the length of the longest data in the binary data of the deviation values ​​of all pixel values ​​in the motion data segment is obtained and recorded as the reference length; and the binary data of the deviation value of each pixel value in the motion data segment is adjusted to the reference length by zero padding, including:

[0073] The difference between each pixel value and the minimum value in the motion data segment is recorded as the deviation value; the length of the longest data in the binary data of the deviation values ​​of all pixel values ​​in the motion data segment is recorded as the reference length; the length of the binary data of the deviation values ​​of the pixel values ​​in the motion data segment is compared with the reference length. If the length of the binary data of the deviation values ​​of the pixel values ​​in the motion data segment is less than the reference length, the length of the binary data of the deviation values ​​of the pixel values ​​in the motion data segment is adjusted to the reference length by adding a number of zeros before the binary data of the deviation values ​​of the pixel values ​​in the motion data segment. If the length of the binary data of the deviation values ​​of the pixel values ​​in the motion data segment is not less than the reference length, no zero padding is required.

[0074] It should be noted that since the data differences in the motion data segments are small, the binary data length differences in the motion data segments are small. By adjusting the binary data in the motion data segments to the same length, the amount of wasted data is also relatively small, thereby effectively reducing the amount of unnecessary data transmission.

[0075] S32: Obtain binary data of the deviation value of each pixel value in the background image, and adjust the binary data of the deviation value of each pixel value in the background image.

[0076] Preferably, as an example, obtaining binary data of the deviation value of each pixel value in the background image and adjusting the binary data of the deviation value of each pixel value in the background image include:

[0077] The sequence of pixel values ​​at a pixel position in the background image of all frame images is recorded as a background time series sequence, and the background time series sequence is divided into a plurality of data segments according to the method of dividing the motion time series sequence into motion data segments in step S30, which are recorded as background data segments;

[0078] The difference between each pixel value and the minimum value in the background data segment is obtained as the deviation value of each pixel value in the background pixel point, the binary data of the deviation value of each pixel value in the background data segment is obtained, and the maximum length of the binary data of all pixel values ​​in the background data segment is used as the reference length of the background data segment, and the binary data of the deviation value of each pixel value in the background data segment is adjusted to the corresponding reference length.

[0079] It should be added that due to the existence of moving images, a pixel position has a pixel value in the background images of some frames, but does not have a pixel value in the background images of some frames. If the pixel value does not exist, it is simply skipped and the sequence consisting of the pixel values ​​of a pixel position in the background images of all frames is used as the background timing sequence.

[0080] It can be understood that since the background objects generally change less in time, the data differences in the background time series composed of the colors of the background objects are smaller. By further segmenting the background time series sequence, the data differences in the background data segments are made smaller. Therefore, by adjusting the binary numbers with smaller data differences to the same length, less unnecessary data is wasted.

[0081] S4: taking the trajectory sequence, the adjusted data of the binary data of the deviation value of each pixel in the motion image block, and the adjusted data of the binary data of the deviation value of each pixel value in the background image as transmission objects to achieve compressed transmission.

[0082] Preferably, as an example, the trajectory sequence, the adjusted data of the binary data of the deviation value of each pixel in the motion image block, and the adjusted data of the binary data of the deviation value of each pixel value in the background image are used as transmission objects to achieve compressed transmission, including:

[0083] The sequence obtained by splicing together the binary adjusted data of the deviation values ​​of all pixel values ​​of each motion data segment is recorded as a unit sequence, and the minimum value of all pixel values ​​and the number of pixel values ​​of the motion data segment are inserted before the unit sequence, and then the preset motion segmentation flag value is inserted; the unit sequences after the motion segmentation flag value is inserted into all motion data segments in the motion timing sequence are spliced ​​together to obtain a first sequence of a pixel position in the motion image; the preset motion pixel flag value is inserted before the first sequence, and the first sequence after the motion pixel flag value is inserted into all pixel positions in the motion image is spliced ​​together to obtain a transmission sequence of the motion image; in this embodiment, the preset motion segmentation flag value is set to A, and the preset motion pixel flag value is set to B. Other embodiments may set other values, and this embodiment does not make specific restrictions.

[0084] The sequence obtained by splicing together the binary adjusted data of the deviation values ​​of all pixel values ​​of each background data segment is recorded as the unit sequence of the background data segment. After inserting the minimum value and the number of pixel values ​​of all pixel values ​​of the background data segment into the unit sequence of the background data segment, the preset background segmentation mark value is inserted; the unit sequence after all background data segments in the background timing sequence are inserted into the background segmentation mark value is spliced ​​together to obtain a second sequence of a pixel position in the background image; the preset background pixel mark value is inserted before the second sequence, and the second sequence after all pixel positions in the background image are inserted into the background pixel mark value is spliced ​​together to obtain the transmission sequence of the background image. In this embodiment, the preset background segmentation mark value is set to , set the preset background pixel flag value to , other embodiments may set other values, and this embodiment does not specifically limit

[0085] Insert the starting frame number and the ending frame number corresponding to the trajectory sequence before the trajectory sequence; splice the trajectory sequence after the inserted frame number, the reference size, the transmission sequence of the motion image and the transmission sequence of the background image together as the transmission sequence of the channel image, take the transmission sequence of the channel image as the transmission object, and perform transmission processing.

[0086] It should be supplemented that the starting frame number and the ending frame number corresponding to the trajectory sequence are the lower limit data and the upper limit data of the frame number interval of the image where the moving object corresponding to the trajectory sequence is located.

[0087] It should be noted that in order to enable subsequent decompression, some necessary information needs to be inserted into the transmitted data to assist in the subsequent decompression process.

[0088] It should be further explained that, except for the pixel value for adjusting the binary length in the transmission sequence, other data adopts the ordinary data placeholder method, such as the length and width of the reference size, which usually adopt a data length of 1 byte, and the 1 byte is the data placeholder.

[0089] S5: Decompress the transmission sequence of the channel images to restore each channel image.

[0090] It should be noted that, for ease of implementation, the decompression process is briefly introduced below.

[0091] Preferably, as an example, decompressing the transmission sequence of the channel images to restore each channel image includes:

[0092] The starting position of the motion image transmission sequence is obtained according to the first motion segmentation mark value in the transmission sequence of the channel image, the reference size is located in the first two data places of the starting position of the motion image transmission sequence, and the trajectory sequence after the inserted frame number is located in the place before the reference size.

[0093] The starting frame number, the ending frame number and the center coordinates of all frame motion images in the trajectory sequence after the inserted frame number are obtained; and the coordinates of each pixel of each frame motion image are obtained according to the reference size and the center coordinates.

[0094] It should be noted that the frame position of the motion image and the coordinates of each pixel in the motion image have been located at this time, and the pixel value of each pixel in each frame of the motion image needs to be decompressed.

[0095] According to the motion pixel flag value in the transmission sequence of the motion image, the first sequence after the motion pixel flag value is inserted at each position in the motion image is obtained, and according to the motion segmentation flag in the first sequence, a unit sequence after several motion data segments are inserted into the motion segmentation flag value is obtained, and the minimum value and the number of pixel values ​​of the motion data segment are located in the second and third data places of the unit sequence after the motion data segment is inserted into the motion segmentation flag value, and the unit sequence is divided into M parts, M represents the number of pixel values, and the data stored in each part is a deviation value, and the deviation value and the minimum value of the motion data segment are added to obtain the pixel value.

[0096] It should be noted that the first sequence describes all deviation values ​​of a position of all frame motion images. Therefore, the deviation value and the minimum value can be used to restore all pixel values ​​of a position of all frame motion images and the motion timing sequence composed of all pixels of a position of all frame motion images.

[0097] In the same way as the motion image, the background time sequence of each pixel position is decompressed; the missing frame segment in the background time sequence is obtained according to the frame position of the motion image and the coordinates of each pixel of the motion image.

[0098] The frame position corresponding to each pixel value in the background time sequence of each pixel position is located according to the missing frame segment in the background time sequence. The decoding method of each pixel value of the background image is the same as the decoding method of each pixel value of each frame of the motion image.

[0099] For example, according to the frame position of the motion image and the coordinates of each pixel of the motion image, the 8th pixel position in the channel image of the 8th to 15th frames belongs to the motion image. There is no pixel value in the background image of the 8th to 15th frames, so the 8th to 15th frames are missing frame segments of the background timing sequence. According to the missing frame segments, it can be known that the pixel value of the 7th frame of the background timing sequence is directly followed by the pixel value of the 16th frame.

[0100] An embodiment of the present invention also discloses a smart medical interaction system based on big data, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a smart medical interaction method based on big data according to the present invention is implemented.

[0101] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, and their configuration and functions are known in the art, so they will not be described in detail here.

[0102] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory, a dynamic random access memory, a static random access memory, an enhanced dynamic random access memory, a high bandwidth memory, a hybrid storage cube, etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of the device or accessible or connectable to the device.

[0103] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

[0104] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A smart medical interaction method based on big data, characterized in that: Includes steps: Obtain the smart medical video and each channel image of each frame image in the smart medical video; Acquire the area where the moving object is located and the area where the non-moving object is located in the channel image obtained in advance, and record them as the moving image and the background image respectively; The sequence constructed by the center coordinates of the motion images in all frame images is recorded as the trajectory sequence; Recording a sequence of pixel values ​​at a pixel position in the running images of all frame images as a motion time series sequence, and dividing the motion time series sequence into a plurality of data segments recorded as motion data segments; The difference between each pixel value in the motion data segment and the minimum value is recorded as a deviation value, and the length of the longest data obtained from the binary data of the deviation values ​​of all pixel values ​​in the motion data segment is recorded as a reference length; The binary data of the deviation value of each pixel value in the motion data segment is adjusted to the reference length by zero padding; the binary data of the deviation value of each pixel value in the background image is obtained, and the binary data of the deviation value of each pixel value in the background image is adjusted; The sequence obtained by splicing together the binary adjusted data of the deviation values ​​of all pixel values ​​of each motion data segment is recorded as a unit sequence, and the minimum value of all pixel values ​​of the motion data segment, the number of pixel values ​​and the preset motion segmentation mark value are inserted before each unit sequence, and then the splicing process is performed to obtain a first sequence of a pixel position in the motion image; the preset motion pixel mark value is inserted before each first sequence, and then the splicing process is performed to obtain a transmission sequence of the motion image; the transmission sequence of the background image is obtained; the starting frame number and the ending frame number corresponding to the trajectory sequence are inserted before the trajectory sequence, and then the reference size, the transmission sequence of the motion image and the transmission sequence of the background image are spliced ​​together as the transmission data sequence of the channel image, and the transmission data sequence is used as a transmission object to achieve compressed transmission.

2. According to claim 1, a smart medical interaction method based on big data is characterized in that: The obtaining of the region where the moving object is located and the region where the non-moving object is located in the channel image obtained in advance are recorded as the moving image and the background image respectively, including: Perform grayscale processing on each frame image to obtain a grayscale image; use a frame difference method to process the grayscale images of all frame images to obtain a frame difference image of each frame image; obtain the bounding rectangle of the area composed of pixels whose grayscale values ​​are not equal to zero in the frame difference image; obtain the size corresponding to the largest bounding rectangle among the bounding rectangles in the frame difference images of all frame images and record it as the reference size; In the channel image, an image area enclosed by a rectangle with a reference size and a center of a circumscribed rectangle in a frame difference image of a corresponding frame as the center is obtained and recorded as a motion image; and an area other than the motion image in the channel image is recorded as a background image.

3. According to the big data-based intelligent medical interaction method of claim 1, it is characterized in that: The sequence constructed by the center coordinates of the motion images in all frame images is recorded as a trajectory sequence, including: In the smart medical video, a frame segment from the beginning of the appearance to the disappearance of a moving object is obtained, the center coordinates of all moving images corresponding to the moving object are obtained in the frame segment, and the sequence formed by them is recorded as a trajectory sequence.

4. The big data-based intelligent medical interaction method according to claim 3, characterized in that: The method for acquiring the frame segment comprises: Acquire an individual region; acquire a reference individual region in a frame difference image of a next frame, and calculate the morphological similarity between the individual region and the reference individual region; multiply the morphological similarity by the inverse of the distance between the individual region and the reference individual region, and then perform normalization processing to obtain the probability of being the same individual; if the probability of being the same individual is greater than a preset probability threshold, it is determined that the moving object corresponding to the individual region and the reference individual region is the same moving object; if the probability of being the same individual is not greater than the preset probability threshold, it is determined that the moving object corresponding to the individual region and the reference individual region is not the same moving object; and acquire the frame number interval corresponding to the image in which the same moving object exists as the frame segment in which the same moving object begins to appear and disappears.

5. The big data-based intelligent medical interaction method according to claim 3, characterized in that: The sequence composed of pixel values ​​at a pixel position in the running images of all frame images is recorded as a motion time sequence, and the motion time sequence is divided into a plurality of data segments recorded as motion data segments, including: Obtain a frame segment from the beginning of the appearance to the disappearance of a moving object in the smart medical video, and record a sequence consisting of all pixel values ​​of a pixel position of all moving images in the frame segment as a motion time sequence; The pixel values ​​in the motion time series are clustered to obtain several categories, and the pixels belonging to the same category in the motion time series are segmented into one data segment to obtain several data segments recorded as motion data segments.

6. The big data-based intelligent medical interaction method according to claim 1, characterized in that: The step of adjusting the binary data of the deviation value of each pixel value in the motion data segment to a reference length by zero padding includes: The length of the binary data of the deviation value of the pixel value in the motion data segment is compared with the reference length. If the length of the binary data of the deviation value of the pixel value in the motion data segment is less than the reference length, the length of the binary data of the deviation value of the pixel value in the motion data segment is adjusted to the reference length by adding several zeros before the binary data of the deviation value of the pixel value in the motion data segment. If the length of the binary data of the deviation value of the pixel value in the motion data segment is not less than the reference length, no zero padding is required.

7. The big data-based intelligent medical interaction method according to claim 1, characterized in that: The step of obtaining the binary data of the deviation value of each pixel value in the background image and adjusting the binary data of the deviation value of each pixel value in the background image includes: The sequence of pixel values ​​at a pixel position in the background image of all frame images is recorded as a background timing sequence, the background timing sequence is divided into several data segments recorded as background data segments, the binary data of the deviation value of each pixel value in the background data segment and the corresponding reference length are obtained, and the binary data of the deviation value of each pixel value in the background data segment is adjusted to the corresponding reference length.

8. A smart medical interaction system based on big data, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a smart medical interaction method based on big data according to any one of claims 1 to 7 is implemented.

Images