Man-machine interaction image data compression strategy customization system

Abstract

The invention relates to a man-machine interaction image data compression strategy customization system. The system comprises a compression execution device, an information capturing device, a sequential upgrading device and an edge analysis device. According to the invention, a certain image compression ratio can be ensured while excessive wireless transmission data in man-machine interaction can be avoided.

Description

Technical Field

[0001] The present invention relates to the field of human-computer interaction, and in particular to a human-computer interaction image data compression strategy customization system. Background Art

[0002] Human-computer interaction refers to the application field of human-machine interaction, which is widely used in modern society. For various instructions of the human body, some are directly input by the user, and some are analyzed by the user's image to receive the user's instructions, which involves image processing issues.

[0003] The invention of application publication number CN116567270A proposes a HZG lossless high-compression image and text encoding method, including multi-scalable wavelet transform processing of original image and text data, so as to perform adaptive adjustment of data transmission quality according to the code stream transmission conditions of different terminals, and on this basis, sample point reconstruction of image and text data in the multi-scalable wavelet transform processing, adaptive syntax element arithmetic encoding and decoding of the reconstructed image and text data, and finally realize lossless high-compression encoding and decoding operation of image and text data. The invention controls real-time compression within 200 milliseconds, with a large compression ratio, and is used to compress 1G of MOV original video to within 30MB, with a compression ratio of up to 30 times, and does not change the resolution and does not lose frames.

[0004] The invention of application publication number CN116567300A proposes a video processing method, device, system and storage medium. The playback end of the local device, with the support of a software decoding program written in a preset programming language, expands the decoding capability of the target video compression standard, and reduces the demand for network bandwidth resources when the video data of the target video compression standard is transmitted between the cloud device and the local device. In this way, the playback end of the local device can decode and play the video data of the target video compression standard with superior performance transmitted by the cloud device. The transmission of video data of the target video compression standard by the cloud device to the playback end of the local device can greatly reduce the consumption of network bandwidth, reduce the delay of the playback end, improve the streaming efficiency of the playback end, better achieve low-latency playback, and improve user experience. Summary of the invention

[0005] According to the present invention, a human-computer interaction image data compression strategy customization system is provided, the system comprising: A compression execution device, for increasing the compression ratio of the real-time live broadcast picture of the current live broadcast room upon receiving a missing registration anchor command, so as to compress the real-time live broadcast picture of the current live broadcast room, and sending the compressed data to a wireless upload device wirelessly connected to the live broadcast server; An information capturing device, used for capturing a preview image of a real-time live broadcast screen of a current live broadcast room, wherein the resolution of the preview image is smaller than the resolution of the live broadcast screen; A sequential quality improvement device, comprising a box filter device, a content enhancement device and a directional denoising device, wherein the box filter device is connected to the information capture device and is used to perform a box filter process on the received preview image to obtain and output a corresponding box filter image, the content enhancement device is connected to the box filter device and is used to perform an image spatial domain enhancement process on the received box filter image to obtain and output a corresponding spatial domain enhanced image, and the directional denoising device is connected to the content enhancement device and is used to perform a Gaussian white noise elimination process on the received spatial domain enhanced image to obtain and output a corresponding directional denoised image; The edge analysis device is connected to the compression execution device and the content enhancement device respectively, and is used to obtain the hue channel value, brightness channel value and saturation channel value of each pixel in the directional denoising image in the HSL space, determine its hue channel gradient value, brightness channel gradient value and saturation channel gradient value based on the hue channel value, brightness channel gradient value and saturation channel gradient value of its adjacent pixels, and when there is a channel gradient value greater than or equal to a set gradient value limit among its hue channel gradient value, brightness channel gradient value and saturation channel gradient value, it is judged to be an edge pixel point, so as to obtain each edge pixel point in the directional denoising image, combine the each edge pixel point to obtain one or more closed areas in the directional denoising image, and when there is a closed area matching the standard human body shape pattern corresponding to the registered anchor in the one or more closed areas, a registered anchor existence command is issued, otherwise, a registered anchor missing command is issued, and the registered anchor is the anchor personnel of the current live broadcast room at the live broadcast server end level.

[0006] The present invention aims to solve the technical problems in the prior art. When a targeted visual detection mechanism is used to detect that there is no registered anchor content in the current live broadcast room in the preview image of the real-time live broadcast picture, the compression ratio of the real-time live broadcast picture of the current live broadcast room can be increased to compress the real-time live broadcast picture of the current live broadcast room, and the compressed data can be sent to a wireless upload device wirelessly connected to a live broadcast server. Otherwise, the compression ratio of the real-time live broadcast picture of the current live broadcast room can be reduced to compress the real-time live broadcast picture of the current live broadcast room, and the compressed data can be sent to a wireless upload device wirelessly connected to the live broadcast server, thereby ensuring a certain image compression ratio while avoiding excessive wireless transmission of data. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] The embodiments of the present invention will be described below with reference to the accompanying drawings.

[0008] Figure 1 The figure is a schematic diagram of the internal structure of the system for customizing the human-computer interaction image data compression strategy according to the first embodiment of the present invention.

[0009] Figure 2 Schematic diagram of the internal structure of a system for customizing human-computer interaction image data compression strategy according to a second embodiment of the present invention.

[0010] Figure 3 Schematic diagram of the internal structure of a system for customizing human-computer interaction image data compression strategy according to a third embodiment of the present invention. DETAILED DESCRIPTION

[0011] Live broadcast operation refers to a series of management and marketing activities conducted on the live broadcast platform, including but not limited to anchor management, content planning, fan interaction, etc. The purpose of live broadcast operation is to improve the user activity and retention rate of the live broadcast platform, attract more viewers and advertisers, and thus realize the commercial value of the platform. On the live broadcast platform, users are the key to the development of the platform. High-quality live broadcast content and good anchor management can improve user retention and activity. Through live broadcast operation, accurate content push and interaction can be carried out according to user needs, enhancing user stickiness to the platform and improving user participation and loyalty.

[0012] However, live broadcast operations cannot just focus on users. The quality of the video and the speed and efficiency of video processing are also critical. In the existing technology, the compression processing of the real-time live broadcast picture of the current live broadcast room does not consider whether there is a host lens in the real-time live broadcast picture of the current live broadcast room. This uniform compression mode causes the entire live broadcast process to consume a large amount of storage space and computing space, affecting the speed and efficiency of live data transmission and playback.

[0013] The embodiment of the human-computer interaction image data compression strategy customization system of the present invention will be described in detail below with reference to the accompanying drawings.

[0014] First embodiment

[0015] Figure 1 This is a schematic diagram of the internal structure of a system for customizing a human-computer interaction image data compression strategy according to a first embodiment of the present invention. The system includes: A compression execution device, for increasing the compression ratio of the real-time live broadcast picture of the current live broadcast room upon receiving a missing registration anchor command, so as to compress the real-time live broadcast picture of the current live broadcast room, and sending the compressed data to a wireless upload device wirelessly connected to the live broadcast server; For example, the compression execution device is used to increase the compression ratio of the real-time live broadcast picture of the current live broadcast room to compress the real-time live broadcast picture of the current live broadcast room when receiving the registered anchor missing command, and send the compressed data to the wireless upload device wirelessly connected to the live broadcast server, including: adopting HEVC compression encoding mode or MPEG compression encoding mode to realize compression processing of the real-time live broadcast picture of the current live broadcast room; An information capturing device, used for capturing a preview image of a real-time live broadcast screen of a current live broadcast room, wherein the resolution of the preview image is smaller than the resolution of the live broadcast screen; A sequential quality improvement device, comprising a box filter device, a content enhancement device and a directional denoising device, wherein the box filter device is connected to the information capture device and is used to perform a box filter process on the received preview image to obtain and output a corresponding box filter image, the content enhancement device is connected to the box filter device and is used to perform an image spatial domain enhancement process on the received box filter image to obtain and output a corresponding spatial domain enhanced image, and the directional denoising device is connected to the content enhancement device and is used to perform a Gaussian white noise elimination process on the received spatial domain enhanced image to obtain and output a corresponding directional denoised image; an edge analysis device, connected to the compression execution device and the content enhancement device respectively, for obtaining the hue channel value, brightness channel value and saturation channel value of each pixel in the directional denoised image in the HSL space, determining its hue channel gradient value, brightness channel gradient value and saturation channel gradient value based on the hue channel value, brightness channel gradient value and saturation channel gradient value of its adjacent pixels, and judging it as an edge pixel when there is a channel gradient value greater than or equal to a set gradient value limit among its hue channel gradient value, brightness channel gradient value and saturation channel gradient value, so as to obtain each edge pixel in the directional denoised image, combining the each edge pixel to obtain one or more closed areas in the directional denoised image, and issuing a registered anchor presence command when there is a closed area matching the standard human body shape pattern corresponding to the registered anchor in the one or more closed areas, otherwise, issuing a registered anchor absence command, the registered anchor is the anchor personnel of the current live broadcast room at the live broadcast server end level; Among them, the compression execution device is also used to reduce the compression ratio of the real-time live broadcast picture of the current live broadcast room when receiving the registration anchor existence command to compress the real-time live broadcast picture of the current live broadcast room, and send the compressed data to the wireless upload device wirelessly connected to the live broadcast server.

[0016] Second embodiment

[0017] Figure 2 Schematic diagram of the internal structure of a system for customizing human-computer interaction image data compression strategy according to a second embodiment of the present invention.

[0018] Figure 2 The human-computer interaction image data compression strategy customization system in the paper can include the following components: A compression execution device, for increasing the compression ratio of the real-time live broadcast picture of the current live broadcast room upon receiving a missing registration anchor command, so as to compress the real-time live broadcast picture of the current live broadcast room, and sending the compressed data to a wireless upload device wirelessly connected to the live broadcast server; An information capturing device, used for capturing a preview image of a real-time live broadcast screen of a current live broadcast room, wherein the resolution of the preview image is smaller than the resolution of the live broadcast screen; A sequential quality improvement device, comprising a box filter device, a content enhancement device and a directional denoising device, wherein the box filter device is connected to the information capture device and is used to perform a box filter process on the received preview image to obtain and output a corresponding box filter image, the content enhancement device is connected to the box filter device and is used to perform an image spatial domain enhancement process on the received box filter image to obtain and output a corresponding spatial domain enhanced image, and the directional denoising device is connected to the content enhancement device and is used to perform a Gaussian white noise elimination process on the received spatial domain enhanced image to obtain and output a corresponding directional denoised image; an edge analysis device, connected to the compression execution device and the content enhancement device respectively, for obtaining the hue channel value, brightness channel value and saturation channel value of each pixel in the directional denoised image in the HSL space, determining its hue channel gradient value, brightness channel gradient value and saturation channel gradient value based on the hue channel value, brightness channel gradient value and saturation channel gradient value of its adjacent pixels, and judging it as an edge pixel when there is a channel gradient value greater than or equal to a set gradient value limit among its hue channel gradient value, brightness channel gradient value and saturation channel gradient value, so as to obtain each edge pixel in the directional denoised image, combining the each edge pixel to obtain one or more closed areas in the directional denoised image, and issuing a registered anchor presence command when there is a closed area matching the standard human body shape pattern corresponding to the registered anchor in the one or more closed areas, otherwise, issuing a registered anchor absence command, the registered anchor is the anchor personnel of the current live broadcast room at the live broadcast server end level; A sequential filtering device, connected to the box filtering device, the content enhancement device and the directional denoising device, respectively, for sequentially filtering the output image signals of the box filtering device, the content enhancement device and the directional denoising device, respectively, to obtain sequentially filtered images corresponding to the output image signals of the box filtering device, the content enhancement device and the directional denoising device, respectively; The step of performing sequential filtering on the output image signals of the box filter device, the content enhancement device and the directional denoising device to obtain sequential filtered images corresponding to the output image signals of the box filter device, the content enhancement device and the directional denoising device respectively comprises: performing homomorphic filtering on the output image signals of the box filter device, the content enhancement device and the directional denoising device respectively in the first layer of filtering; The method of performing sequential filtering on the output image signals of the box filter device, the content enhancement device and the directional denoising device to obtain the sequential filtered images corresponding to the output image signals of the box filter device, the content enhancement device and the directional denoising device respectively further comprises: performing bandpass filtering on the output image signals of the box filter device, the content enhancement device and the directional denoising device respectively in the second layer filtering; The method of performing sequential filtering on the output image signals of the box filter device, the content enhancement device and the directional denoising device to obtain the sequential filtered images corresponding to the output image signals of the box filter device, the content enhancement device and the directional denoising device respectively further comprises: performing midpoint filtering on the output image signals of the box filter device, the content enhancement device and the directional denoising device respectively in a third layer of filtering; And wherein, the respective output image signals of the box filter device, the content enhancement device and the directional denoising device are sequentially filtered to obtain the respective sequential filtered images corresponding to the respective output image signals of the box filter device, the content enhancement device and the directional denoising device, and further includes: the fourth layer filtering performs median filtering processing on the respective output image signals of the box filter device, the content enhancement device and the directional denoising device.

[0019] Third embodiment

[0020] Figure 3 Schematic diagram of the internal structure of a system for customizing human-computer interaction image data compression strategy according to a third embodiment of the present invention.

[0021] Figure 3 The human-computer interaction image data compression strategy customization system in the paper can include the following components: A compression execution device, for increasing the compression ratio of the real-time live broadcast picture of the current live broadcast room upon receiving a missing registration anchor command, so as to compress the real-time live broadcast picture of the current live broadcast room, and sending the compressed data to a wireless upload device wirelessly connected to the live broadcast server; An information capturing device, used for capturing a preview image of a real-time live broadcast screen of a current live broadcast room, wherein the resolution of the preview image is smaller than the resolution of the live broadcast screen; A sequential quality improvement device, comprising a box filter device, a content enhancement device and a directional denoising device, wherein the box filter device is connected to the information capture device and is used to perform a box filter process on the received preview image to obtain and output a corresponding box filter image, the content enhancement device is connected to the box filter device and is used to perform an image spatial domain enhancement process on the received box filter image to obtain and output a corresponding spatial domain enhanced image, and the directional denoising device is connected to the content enhancement device and is used to perform a Gaussian white noise elimination process on the received spatial domain enhanced image to obtain and output a corresponding directional denoised image; an edge analysis device, connected to the compression execution device and the content enhancement device respectively, for obtaining the hue channel value, brightness channel value and saturation channel value of each pixel in the directional denoised image in the HSL space, determining its hue channel gradient value, brightness channel gradient value and saturation channel gradient value based on the hue channel value, brightness channel gradient value and saturation channel gradient value of its adjacent pixels, and judging it as an edge pixel when there is a channel gradient value greater than or equal to a set gradient value limit among its hue channel gradient value, brightness channel gradient value and saturation channel gradient value, so as to obtain each edge pixel in the directional denoised image, combining the each edge pixel to obtain one or more closed areas in the directional denoised image, and issuing a registered anchor presence command when there is a closed area matching the standard human body shape pattern corresponding to the registered anchor in the one or more closed areas, otherwise, issuing a registered anchor absence command, the registered anchor is the anchor personnel of the current live broadcast room at the live broadcast server end level; A humidity measuring device, connected to the box-type filtering device, the content enhancement device and the directional denoising device, respectively, for performing on-site measurement of the internal humidity of each of the box-type filtering device, the content enhancement device and the directional denoising device; The humidity measuring device is connected to the box-type filter device, the content enhancement device and the directional denoising device, respectively, and is used to perform on-site measurement of the internal humidity of the box-type filter device, the content enhancement device and the directional denoising device, respectively, including: the humidity measuring device uses three humidity measuring units, which are connected to the box-type filter device, the content enhancement device and the directional denoising device, respectively, and are used to perform on-site measurement of the internal humidity of the box-type filter device, the content enhancement device and the directional denoising device, respectively; And wherein, the humidity measuring device uses three humidity measuring units, which are respectively connected to the box-type filtering device, the content enhancement device and the directional denoising device, for respectively performing on-site measurement of the internal humidity of the box-type filtering device, the content enhancement device and the directional denoising device: the measurement ranges of the three humidity measuring units are the same and the internal structures of the three humidity measuring units are the same.

[0022] In addition, in the human-computer interaction image data compression strategy customization system, the hue channel value, brightness channel value and saturation channel value of each pixel point in the directional denoising image in the HSL space are obtained, and the hue channel gradient value, brightness channel gradient value and saturation channel gradient value of the pixel point are determined based on the hue channel value, brightness channel gradient value and saturation channel gradient value of the pixel point that is adjacent to the hue channel value, brightness channel gradient value and saturation channel gradient value, and when there is a channel gradient value greater than or equal to the set gradient value limit among the hue channel gradient value, brightness channel gradient value and saturation channel gradient value, it is judged to be an edge pixel point, including: the value of any channel value of the hue channel value, brightness channel value and saturation channel value of each pixel point in the directional denoising image in the HSL space is between 0 and 255.

[0023] The technical innovations of the present invention are as follows: (1) When it is detected by a targeted visual detection mechanism that there is no registered anchor content of the current live broadcast room in the preview image of the real-time live broadcast picture, the compression ratio of the real-time live broadcast picture of the current live broadcast room is increased to compress the real-time live broadcast picture of the current live broadcast room, and the compressed data is sent to a wireless upload device wirelessly connected to the live broadcast server; otherwise, the compression ratio of the real-time live broadcast picture of the current live broadcast room is reduced to compress the real-time live broadcast picture of the current live broadcast room, and the compressed data is sent to a wireless upload device wirelessly connected to the live broadcast server, thereby ensuring a certain image compression ratio while avoiding excessive wireless transmission of data; (2) Obtain the hue channel value, brightness channel value and saturation channel value of each pixel in the directional denoised image in the HSL space, determine its hue channel gradient value, brightness channel gradient value and saturation channel gradient value based on the hue channel value, brightness channel gradient value and saturation channel gradient value of its adjacent pixels, and when there is a channel gradient value greater than or equal to the set gradient value limit among its hue channel gradient value, brightness channel gradient value and saturation channel gradient value, determine it as an edge pixel, so as to obtain each edge pixel in the directional denoised image; (3) The edge pixels in the directional denoised image are combined to obtain one or more closed areas in the directional denoised image. When there is a closed area matching the standard human body shape pattern corresponding to the registered anchor in the one or more closed areas, a registered anchor existence command is issued, thereby reducing the detection area of ​​the registered anchor and improving the detection efficiency of the registered anchor.

[0024] The human-computer interaction image data compression strategy customization system of the present invention can solve the technical problem in the prior art that real-time live broadcast images are compressed indiscriminately, resulting in huge storage costs and computing costs. When it is detected that the registered anchor content of the current live broadcast room does not exist in the preview image of the real-time live broadcast image, the compression ratio of the real-time live broadcast image can be improved for compression, and the compressed data can be sent to the live broadcast server, thereby ensuring a certain image compression ratio while avoiding excessive wireless transmission of data.

[0025] Although some embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the claims and their equivalents.

Claims

1. A human-computer interactive image data compression strategy customization system, characterized in that: The system comprises: A compression execution device, for increasing the compression ratio of the real-time live broadcast picture of the current live broadcast room upon receiving a missing registration anchor command, so as to compress the real-time live broadcast picture of the current live broadcast room, and sending the compressed data to a wireless upload device wirelessly connected to the live broadcast server; An information capture device, used to capture a preview image of a real-time live broadcast screen of a current live broadcast room, wherein the resolution of the preview image is smaller than the resolution of the live broadcast screen; A sequential quality improvement device, comprising a box filter device, a content enhancement device and a directional denoising device, wherein the box filter device is connected to the information capture device and is used to perform a box filter process on the received preview image to obtain and output a corresponding box filter image, the content enhancement device is connected to the box filter device and is used to perform an image spatial domain enhancement process on the received box filter image to obtain and output a corresponding spatial domain enhanced image, and the directional denoising device is connected to the content enhancement device and is used to perform a Gaussian white noise elimination process on the received spatial domain enhanced image to obtain and output a corresponding directional denoised image; The edge analysis device is connected to the compression execution device and the content enhancement device respectively, and is used to obtain the hue channel value, brightness channel value and saturation channel value of each pixel in the directional denoising image in the HSL space, determine its hue channel gradient value, brightness channel gradient value and saturation channel gradient value based on the hue channel value, brightness channel gradient value and saturation channel gradient value of its adjacent pixels, and when there is a channel gradient value greater than or equal to a set gradient value limit among its hue channel gradient value, brightness channel gradient value and saturation channel gradient value, it is judged to be an edge pixel point, so as to obtain each edge pixel point in the directional denoising image, combine the each edge pixel point to obtain one or more closed areas in the directional denoising image, and when there is a closed area matching the standard human body shape pattern corresponding to the registered anchor in the one or more closed areas, a registered anchor existence command is issued, otherwise, a registered anchor missing command is issued, and the registered anchor is the anchor personnel of the current live broadcast room at the live broadcast server end level.

2. The human-computer interaction image data compression strategy customization system according to claim 1, characterized in that: The compression execution device is also used to reduce the compression ratio of the real-time live broadcast picture of the current live broadcast room when receiving the registration anchor existence command to compress the real-time live broadcast picture of the current live broadcast room, and send the compressed data to the wireless upload device wirelessly connected to the live broadcast server.

3. The human-computer interaction image data compression strategy customization system according to claim 2, characterized in that: The system further comprises: A sequential filtering device, connected to the box filtering device, the content enhancement device and the directional denoising device, respectively, for sequentially filtering the output image signals of the box filtering device, the content enhancement device and the directional denoising device, respectively, to obtain sequentially filtered images corresponding to the output image signals of the box filtering device, the content enhancement device and the directional denoising device, respectively; Among them, the output image signals of the box filter device, the content enhancement device and the directional denoising device are respectively sequentially filtered to obtain the sequential filtered images corresponding to the output image signals of the box filter device, the content enhancement device and the directional denoising device, respectively, including: the first layer filtering performs homomorphic filtering processing on the output image signals of the box filter device, the content enhancement device and the directional denoising device, respectively.

4. The human-computer interaction image data compression strategy customization system according to claim 3, characterized in that: The respective output image signals of the box filter device, the content enhancement device and the directional denoising device are sequentially filtered to obtain respective sequential filtered images corresponding to the respective output image signals of the box filter device, the content enhancement device and the directional denoising device, and also include: the second layer filtering performs bandpass filtering processing on the respective output image signals of the box filter device, the content enhancement device and the directional denoising device.

5. The human-computer interaction image data compression strategy customization system according to claim 4, characterized in that: The respective output image signals of the box filter device, the content enhancement device and the directional denoising device are sequentially filtered to obtain respective sequential filtered images corresponding to the respective output image signals of the box filter device, the content enhancement device and the directional denoising device, and also include: the third layer filtering performs midpoint filtering processing on the respective output image signals of the box filter device, the content enhancement device and the directional denoising device.

6. The human-computer interaction image data compression strategy customization system according to claim 5, characterized in that: The respective output image signals of the box filter device, the content enhancement device and the directional denoising device are sequentially filtered to obtain respective sequential filtered images corresponding to the respective output image signals of the box filter device, the content enhancement device and the directional denoising device, and also include: the fourth layer filtering performs median filtering processing on the respective output image signals of the box filter device, the content enhancement device and the directional denoising device.

7. The human-computer interaction image data compression strategy customization system according to any one of claims 3 to 6, characterized in that: The system further comprises: The humidity measuring device is connected to the box-type filtering device, the content enhancing device and the directional denoising device respectively, and is used to perform on-site measurement of the internal humidity of the box-type filtering device, the content enhancing device and the directional denoising device respectively.

8. The human-computer interaction image data compression strategy customization system according to claim 7, characterized in that: The humidity measuring device is respectively connected to the box-type filtering device, the content enhancing device and the directional denoising device, and is used to respectively perform on-site measurements of the internal humidity of the box-type filtering device, the content enhancing device and the directional denoising device. The humidity measuring device includes: the humidity measuring device uses three humidity measuring units, which are respectively connected to the box-type filtering device, the content enhancing device and the directional denoising device, and are used to respectively perform on-site measurements of the internal humidity of the box-type filtering device, the content enhancing device and the directional denoising device.

9. The human-computer interaction image data compression strategy customization system according to claim 8, characterized in that: The humidity measuring device uses three humidity measuring units, which are respectively connected to the box filtering device, the content enhancement device and the directional denoising device, and are used to perform on-site measurements of the internal humidity of the box filtering device, the content enhancement device and the directional denoising device respectively: the measurement ranges of the three humidity measuring units are the same and the internal structures of the three humidity measuring units are the same.

Images