Ultrasonic image processing system and method based on H.265 video coding format
The ultrasonic image processing system with H.265 encoding and intelligent parameter adjustment addresses the issue of information loss and artifacts in ultrasound images, ensuring high-quality compression and accurate medical diagnosis.
Patent Information
- Application Number
- CN202411146212.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-20
- Publication Date
- 2024-12-06
AI Technical Summary
Conventional video compression algorithms for ultrasound images, such as H.264 and H.265, result in information loss and artifacts at high compression rates, degrading image quality and affecting diagnostic accuracy.
An ultrasonic image processing system utilizing H.265 video encoding format with preprocessing modules for denoising and enhancement, followed by intelligent parameter adjustment based on image features, ensuring efficient compression without loss of medical information.
The system achieves high-quality ultrasound image compression maintaining diagnostic integrity by dynamically adjusting encoding parameters based on image features, reducing artifacts and information loss.
Smart Images

Figure CN119094785A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of ultrasonic processing technology, and in particular to an ultrasonic image processing system and method based on the H.265 video encoding format. Background Art
[0002] In the field of medical imaging, ultrasound imaging, as a non-invasive, low-cost and real-time detection method, has become an indispensable tool for clinical diagnosis. From cardiovascular examinations to abdominal organ assessments, ultrasound imaging is widely used in a variety of clinical scenarios, providing doctors with intuitive and dynamic imaging information.
[0003] With the explosive growth of medical data and the increasing demand for telemedicine, higher requirements are placed on the high-quality compression and efficient transmission of ultrasound image data. In the prior art, video compression algorithms such as H.264 (AVC) and H.265 (HEVC) are often used to compress ultrasound image data. However, these video compression algorithms are lossy compression algorithms. When compressing ultrasound image data, ultrasound images are prone to information loss and artifacts at high compression rates due to their non-uniformity and unique noise characteristics, resulting in significant image quality degradation, thereby affecting the accuracy of diagnosis.
[0004] Traditional compression algorithms have difficulty achieving efficient compression while maintaining important medical information. Summary of the invention
[0005] In order to solve the above technical problems, the present invention provides an ultrasonic image processing system and method based on the H.265 video encoding format.
[0006] The technical problem solved by the present invention can be achieved by adopting the following technical solutions:
[0007] An ultrasonic image processing system based on H.265 video encoding format, comprising:
[0008] An ultrasonic data processing module, used for preprocessing the received ultrasonic image data to obtain preprocessed ultrasonic image data; wherein the ultrasonic image data is stored in a first predetermined format;
[0009] An image encoding module is connected to the ultrasound data processing module and is used to adjust encoding parameters according to image features of the preprocessed ultrasound image data to encode first data having a second predetermined format; the second predetermined format is an H.265 video encoding format.
[0010] Preferably, the ultrasound data processing module comprises:
[0011] a data acquisition unit, configured to acquire ultrasonic signals received by the ultrasonic probe, digitally process the ultrasonic signals, and generate the ultrasonic image data stored in the first predetermined format;
[0012] The image preprocessing unit is connected to the data acquisition unit and is used to preprocess the ultrasound image data to obtain the preprocessed ultrasound image data.
[0013] Preferably, the first predetermined format is a Digital Imaging and Communications in Medicine format.
[0014] Preferably, the image encoding module comprises:
[0015] An analysis unit, configured to perform image analysis on the preprocessed ultrasound image data to obtain an image analysis result; the image analysis result includes feature information of different feature types;
[0016] an image division unit, connected to the analysis unit, and configured to divide the preprocessed ultrasound image data into a plurality of key areas according to feature information in the image analysis result;
[0017] A parameter adjustment unit, connected to the image division unit, configured to adjust the encoding parameters of the encoding unit according to the feature type of each key area; each key area corresponds to one encoding unit;
[0018] An image encoding unit is connected to the parameter adjustment unit and is used to perform image encoding on the preprocessed ultrasound image data according to the adjusted encoding parameters to obtain the first data.
[0019] Preferably, the parameter adjustment unit is used to adjust the encoding parameters according to the feature type of each key area based on the corresponding relationship between the feature type and the encoding parameter.
[0020] Preferably, the feature types of the image features include image noise features and / or image texture features;
[0021] The coding parameters include quantization parameters and prediction coding parameters. The prediction coding parameters include intra-frame prediction mode and inter-frame prediction mode. The intra-frame prediction mode includes a plane mode, a direct current mode and an angle mode.
[0022] Preferably, it also includes:
[0023] A quality assessment module, connected to the image encoding module, configured to perform quality assessment on the first data to obtain a quality assessment result;
[0024] The image encoding module is further configured to readjust the encoding parameters when the quality assessment result indicates that the quality does not meet the standard.
[0025] Preferably, it also includes:
[0026] A data transmission module, connected to the image encoding module, and configured to transmit the first data having the second predetermined format to a receiving end via wireless transmission;
[0027] The receiving end is used to decode the first data and visually display the decoded first data.
[0028] Preferably, the receiving end is further used to store the first data and / or the decoded first data;
[0029] The receiving end includes a remote device and / or a cloud server.
[0030] The present invention also provides an ultrasonic image processing method based on the H.265 video coding format, which is applied to the ultrasonic image processing system as described above, and the method includes:
[0031] Preprocessing the received ultrasonic image data to obtain preprocessed ultrasonic image data; wherein the ultrasonic image data is stored in a first predetermined format;
[0032] The encoding parameters are adjusted according to the image features of the preprocessed ultrasound image data to encode the first data in a second predetermined format; the second predetermined format is the H.265 video encoding format.
[0033] The advantages or beneficial effects of the technical solution of the present invention are:
[0034] The present invention adjusts the encoding parameters of H.265 video encoding according to the image features of ultrasonic image data, thereby achieving efficient compression of ultrasonic image data while avoiding ultrasonic information loss and artifact generation, improving the quality of ultrasonic images, and providing more accurate and reliable image basis for medical diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS
[0035] Figure 1 This is a structural block diagram of an ultrasonic image processing system based on the H.265 video encoding format in a preferred embodiment of the present invention;
[0036] Figure 2 The figure is a flow chart of an ultrasonic image processing method based on the H.265 video encoding format in a preferred embodiment of the present invention. DETAILED DESCRIPTION
[0037] 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 only 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0038] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.
[0039] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, but they are not intended to limit the present invention.
[0040] See also Figure 1 In a preferred embodiment of the present invention, based on the above problems existing in the prior art, an ultrasound image processing system based on the H.265 video encoding format is provided, comprising:
[0041] Ultrasonic data processing module 1, used for preprocessing received ultrasonic image data to obtain preprocessed ultrasonic image data; wherein the ultrasonic image data is stored in a first predetermined format;
[0042] The image encoding module 2 is connected to the ultrasound data processing module 1 and is used to adjust the encoding parameters according to the image features of the preprocessed ultrasound image data to encode the first data in a second predetermined format; the second predetermined format is the H.265 video encoding format.
[0043] Specifically, in view of the problem that the conventional video compression algorithm used in the prior art to compress ultrasound images is prone to information loss and artifacts at high compression rates, resulting in reduced image quality, and it is difficult to achieve efficient compression while retaining important medical information. In this embodiment, before image encoding, the ultrasound image data is first preprocessed by the ultrasound data processing module 1 to optimize the image quality.
[0044] The preprocessing steps include, but are not limited to, image denoising, edge enhancement, and format conversion. Specifically, for the common speckle noise and shot noise in ultrasound images, an adaptive filtering algorithm is used for denoising to reduce artifacts in the subsequent encoding process; image enhancement improves image clarity by adjusting the brightness and contrast of the image. Then, the preprocessed ultrasound image data is converted into an intermediate format suitable for subsequent encoding processing to ensure data consistency and processability.
[0045] Image encoding module 2 is an H.265 encoding module. After preprocessing, the preprocessed ultrasound image data is encoded with H.265 to compress the data volume. During the H.265 encoding process, image encoding module 2 can intelligently analyze image features and dynamically encode parameters based on them to find the best balance between compression rate and image quality, which not only achieves significant compression of data volume, but also ensures the integrity of medical information and the diagnosability of images.
[0046] As a preferred embodiment, the ultrasonic data processing module 1 includes:
[0047] The data acquisition unit 11 is used to collect the ultrasonic signal received by the ultrasonic probe 10, digitize the ultrasonic signal, and generate ultrasonic image data stored in a first predetermined format;
[0048] The image preprocessing unit 12 is connected to the data acquisition unit 11 and is used to preprocess the ultrasound image data to obtain preprocessed ultrasound image data.
[0049] Specifically, first initialize the handheld ultrasound device and configure device parameters such as frame rate, resolution, and gain. The frame rate determines how many frames are captured per second, the resolution affects the clarity of the image, and the gain adjusts the brightness of the image. After initialization is complete, test the stability of the connection and communication. This is usually done through the diagnostic tools or initialization scripts provided by the device to ensure that the device does not experience connection interruptions or data loss during the acquisition process.
[0050] After starting the device, the data acquisition process begins. During the acquisition process, the ultrasonic probe 10 transmits ultrasonic signals into the human body, receives ultrasonic echo signals reflected from the human tissue interface, and then converts the received analog echo signals into digital signals. This can be achieved through an analog-to-digital converter (ADC) to convert continuous analog signals into discrete digital signals. The intensity, time delay and other information of these echo signals can be processed in a complex manner to generate original ultrasonic image data reflecting the internal structure of the human body.
[0051] During the data collection process, monitor the image quality and device status in real time. For example, detect abnormal conditions such as noise, high or low brightness in the collected image. If there are any abnormalities, adjust the device parameters in time or stop the collection for investigation. You can write a real-time monitoring script to regularly check the data quality and generate reports. Use simple image processing functions such as mean and variance to detect abnormal changes in image quality and issue alarms in time. If the device connection is interrupted or the data reading error is found, immediately process and restore the connection, such as restarting the device or reloading the driver. In addition, set up a suitable retry mechanism to automatically re-collect missing data, reduce human intervention, and ensure data integrity.
[0052] After obtaining the digital signal, the ultrasonic image data is denoised by the image preprocessing unit 12 to avoid the reduction of image quality due to interference from various noises such as speckle noise and shot noise, and to improve image quality and clarity. Denoising can be implemented using a denoising algorithm, including but not limited to adaptive filtering algorithm, Gaussian filtering, median filtering, wavelet transform denoising and bilateral filtering, to eliminate or reduce the influence of noise and improve image clarity and contrast. For example, Gaussian filtering reduces random noise by constructing a blur kernel and smoothing the image, which is achieved through convolution operation. For example, cv2.GaussianBlur(img,(5,5),0) in OpenCV can be used to perform Gaussian filtering on the image. Gaussian filtering blurs the details of the image, so it is necessary to find a balance between reducing noise and retaining image details. For situations where the requirements for detail preservation are high, bilateral filtering can be used, which not only considers the spatial distance but also the pixel intensity difference to achieve the effect of edge denoising.
[0053] Image enhancement includes but is not limited to gamma correction and histogram equalization. Gamma correction adjusts the brightness of an image to make dark and bright details more obvious. By adjusting the gamma value, nonlinear brightness transformation can be achieved. For example, this can be achieved using the OpenCV function cv2.LUT(img,table), where table is a pre-calculated gamma correction lookup table. Histogram equalization is an effective way to adjust image contrast. By equalizing the image histogram, the brightness distribution of the image is made more uniform. OpenCV's cv2.equalizeHist(gray) function can directly perform histogram equalization on grayscale images.
[0054] In order to facilitate data storage and management, the pre-processed ultrasound image data is formatted and stored according to the DICOM standard. The DICOM standard specifies the format, coding, identification and other information of medical image data, ensuring the standardization and compatibility of medical image data and facilitating data transmission and sharing between different medical devices.
[0055] As a preferred implementation, the first predetermined format is a Digital Imaging and Communications in Medicine (DICOM) format.
[0056] As a preferred embodiment, the image encoding module 2 includes:
[0057] The analysis unit 21 is used to perform image analysis on the preprocessed ultrasound image data to obtain image analysis results; the image analysis results include feature information of different feature types;
[0058] An image division unit 22, connected to the analysis unit 21, is used to divide the pre-processed ultrasound image data into a plurality of key areas according to feature information in the image analysis result;
[0059] A parameter adjustment unit 23, connected to the image division unit 22, is used to adjust the encoding parameters of the encoding unit according to the feature type of each key area; each key area corresponds to one encoding unit;
[0060] The image encoding unit 24 is connected to the parameter adjustment unit 23 and is configured to perform image encoding on the preprocessed ultrasound image data according to the adjusted encoding parameters to obtain first data.
[0061] Specifically, the image analysis algorithm integrated in the analysis unit 21 is used, including but not limited to the Canny operator-based edge detection technology to outline the tissue boundary in the image, and the gray level co-occurrence matrix (GLCM) method is used to evaluate the texture complexity to distinguish the texture characteristics of different tissue regions. Through the comprehensive use of multiple algorithms, it is possible to automatically identify the noise area in the image (such as low signal-to-noise ratio or abnormal highlight area), mark the texture complex area (these areas may correspond to specific tissue types or lesions), and integrate these analysis results into a feature information set containing different feature types (such as edge features, texture features, noise features, etc.).
[0062] The image segmentation unit 22 divides the preprocessed ultrasound image data into multiple key areas based on image features, including but not limited to static feature areas and dynamic feature areas. It can also be subdivided into background areas with less noise and simple texture, noise areas, feature gradient change areas, and tissue areas with complex textures and rich diagnostic information.
[0063] Furthermore, the coding parameters include but are not limited to quantization parameters QP and predictive coding parameters.
[0064] The quantization parameter QP is a key parameter for controlling video quality and compression rate. The smaller the QP value, the higher the video quality, but the lower the compression rate; the larger the QP value, the lower the video quality, but the higher the compression rate.
[0065] For ultrasound images, the QP value can be adjusted according to the importance and detail of the image. For example, for images in critical diagnostic areas, the QP value can be appropriately lowered to ensure image quality; for non-critical areas or background images, the QP value can be appropriately increased to save storage space. For example, for noise areas, the quantization parameter QP can be increased to effectively compress data while reducing storage space.
[0066] Furthermore, using a custom quantization table QT can better balance video quality and compression rate. According to the texture characteristics of the ultrasound image, different quantization intervals can be set for different areas. By quantizing the quantization interval of the QB fine-tuning coefficient, the QB value can be adjusted appropriately as needed.
[0067] Prediction coding parameters include intra prediction mode and inter prediction mode.
[0068] Intra-frame prediction uses the coded pixels in the current frame to predict the current pixel. H.265 provides 35 intra-frame prediction modes, including Planar mode, DC mode, and 33 angle modes. For example, for flat areas in ultrasound images, you can choose the DC mode; for areas with rich textures, you can choose the angle mode; for gradient areas, you can choose the Planar mode.
[0069] Inter-frame prediction uses the information of adjacent frames to predict the current frame. For motion areas in ultrasound images (although the motion in ultrasound images is usually small), inter-frame prediction can significantly improve compression efficiency. It is based on motion estimation algorithms and matching criteria (such as SAD, SATD, etc.) to find the best matching block and reduce the residual.
[0070] During the encoding process, intra-frame prediction and inter-frame prediction are used in a balanced manner according to the specific content and motion of the ultrasound image. For example, intra-frame prediction can be used for static or slightly changing areas, while inter-frame prediction can be used for moving areas or areas with large changes.
[0071] The parameter adjustment unit 23 intelligently sets differentiated encoding parameters according to the characteristics of these regions.
[0072] The image coding unit 24 includes a plurality of coding units (CUs), each of which includes a plurality of coding blocks (CBs) and prediction blocks (PBs). Specifically, intra-frame or inter-frame prediction is performed on each CU to generate a prediction value and a residual; the residual is transformed (such as DCT transformation), and then the transformation coefficient is quantized; the quantized transformation coefficient, motion information, prediction information, etc. are entropy encoded to form a compressed video code stream; the compressed video code stream is encapsulated into a first data stream in the H.265 video coding format with a high-quality compression ratio.
[0073] As a preferred implementation, the parameter adjustment unit 23 is used to adjust the encoding parameters according to the feature type of each key area and based on the corresponding relationship between the feature type and the encoding parameter.
[0074] Specifically, a corresponding relationship is pre-set to dynamically adjust the coding parameters of each coding unit, and the corresponding relationship covers the basic rules between different feature types (such as noise characteristics, texture complexity, etc.) and coding parameters (such as quantization parameters, intra-frame prediction modes, etc.).
[0075] Furthermore, the corresponding relationship can be obtained through experiments, and the corresponding relationship between various feature types and optimal encoding parameters can be determined by comparing and analyzing the experimental results. When the parameter adjustment unit 23 receives the key areas and their feature types divided by the image division unit 22, it selects the most appropriate encoding parameters for each area by consulting the corresponding relationship table to ensure that the image encoding process is both efficient and maintains high-quality image information.
[0076] Furthermore, this correspondence can be realized based on a machine learning model. Through the training of a large amount of ultrasound image data in advance, the complex mapping relationship between feature types and optimal encoding parameters can be captured. During the model application process, the most suitable encoding parameters can be automatically selected and adjusted for the specific features of each key area, thereby achieving efficient data compression while maintaining image quality.
[0077] As a preferred implementation, the feature type of the image feature includes image noise feature and / or image texture feature;
[0078] The coding parameters include quantization parameters and prediction coding parameters. The prediction coding parameters include intra-frame prediction mode and inter-frame prediction mode. The intra-frame prediction mode includes planar mode, direct current mode and angle mode.
[0079] Furthermore, before performing H.265 encoding, select the appropriate encoder and tool. Common encoding tools include FFmpeg, NVIDIA NVENC, etc. FFmpeg is a very popular open source multimedia framework that supports a variety of video processing functions. NVIDIA NVENC uses GPU acceleration technology to greatly improve encoding efficiency. After selection, configure the encoder to ensure its normal operation. Configure the FFmpeg environment variables to make it available in the command line, and use commands such as ffmpeg-iinput.mp4-c:vhevc_nvenc output.mp4 to encode. When choosing GPU encoding, you must also ensure that the image data is transferred to the GPU memory.
[0080] Improve video quality and reduce data volume by optimizing encoding configuration. Encoding parameters can also include resolution, frame rate, and bit rate. Set appropriate resolution and frame rate to match the quality of the preprocessed image. H.265 can maintain high video quality at low bit rates. The bit rate determines the degree of video compression and transmission efficiency. When encoding in constant rate factor (CRF) mode, setting an appropriate CRF value (such as 23) can find a balance between quality and size. When using FFmpeg for H.265 encoding, detailed configuration can be performed through command line parameters, such as ffmpeg-iinput.mp4-c:v libx265-preset medium-crf23 output.mp4. For further optimization, using more efficient presets (such as fast, medium, slow) can find a suitable balance between encoding speed and quality. Monitor the encoding process to ensure that system resources are used effectively, and adjust multi-threading parameters to improve encoding efficiency.
[0081] After configuring the encoding parameters, perform the actual encoding operation. Start the encoding process to pass the preprocessed image data to the encoder for compression to generate an H.265 video file. The encoding process may consume a lot of computing resources, so you need to monitor the system resource usage to ensure smooth encoding. The output video file should match the set resolution, frame rate, and bit rate. The encoded video file can be verified by playback software or tools to ensure that there is no obvious quality loss or compression distortion. If problems are found, you need to readjust the encoding parameters and encode again. The final output video file is used for subsequent data transmission to maintain its efficiency and integrity.
[0082] As a preferred embodiment, the invention further comprises:
[0083] The quality assessment module 3 is connected to the image encoding module 2 and is used to perform quality assessment on the first data to obtain a quality assessment result;
[0084] The image encoding module 2 is further configured to readjust encoding parameters when the quality assessment result indicates that the quality does not meet the standards.
[0085] Specifically, the quality assessment module 3 performs quality assessment on the encoded data. The quality assessment process can be implemented based on a machine learning model to determine whether the quality of the encoded image meets a preset standard.
[0086] If the quality does not meet the standard, it is fed back to the image encoding module 2 to readjust the encoding parameters and re-encode until the quality meets the standard.
[0087] Furthermore, the image encoding module 2 outputs first data when the quality assessment result indicates that the quality meets the standard.
[0088] As a preferred embodiment, the invention further comprises:
[0089] The data transmission module 4 is connected to the image encoding module 2 and is used to transmit the first data in the second predetermined format to the receiving end via wireless;
[0090] The receiving end 5 is used to decode the first data and visually display the decoded first data.
[0091] Specifically, the data transmission module 4 is used to transmit the H.265 encoded data that meets the quality standards to the receiving end via wireless methods such as Wi-Fi, 4G / 5G, etc.
[0092] Furthermore, the data transmission module 4 can use transmission protocols such as TCP, UDP, and QUIC. The TCP transmission control protocol provides reliable data transmission, ensuring that the receiving end can receive data in order and handle packet loss, but the transmission delay is relatively high; the UDP user datagram protocol has low latency, but may lose packets and be disordered; QUIC is a modern protocol that combines UDP low latency and TCP reliability, suitable for real-time video transmission. The transmission protocol should be selected according to the actual network conditions and transmission requirements. For example, in a stable internal network environment, TCP can be used; in a real-time streaming media transmission scenario, QUIC is selected.
[0093] When establishing a transmission connection, network configuration and port opening are required on both sides (sender and receiver). According to the selected transmission protocol, configure the corresponding network ports and firewall rules to ensure that the data can pass smoothly. Taking QUIC as an example, the communication of the protocol can be achieved using open source libraries such as QUIC-GO or ngHTTP3. First, the sender initializes the connection parameters and establishes a session with the receiver to ensure a smooth handshake process, including encryption and security verification. After the connection is established, a small amount of test data is initially transmitted to ensure the stability of the link. The network bandwidth test tool can help understand the current network situation and adjust the transmission parameters to achieve the best transmission efficiency.
[0094] After the transmission connection is successfully established, the actual data transmission begins. The encoded H.265 video stream is divided into packets of appropriate size and sent to the receiving end through the network. The use of streaming media transmission technology, such as Real-Time Streaming Protocol (RTSP) or Dynamic Adaptive Streaming Technology (DASH), can achieve stable and smooth transmission. During the transmission process, use traffic monitoring tools (such as Wireshark or Iperf) to monitor the real-time status of data transmission and capture packet loss rate, latency, and bandwidth usage. If transmission anomalies are detected, such as high packet loss rate or network jitter, the corresponding retransmission mechanism and error handling measures need to be triggered to ensure that the data is transmitted to the destination intact and reliably.
[0095] The receiving end needs to be prepared to receive and process the transmitted data. First, listen to a specific port to receive the video data stream. Create a stable receiving environment, such as setting up a server or using a specific receiving terminal device, and ensure that the port is open by configuring a firewall policy. The received data packets are temporarily stored in the memory, or stored to the hard disk for subsequent processing according to specific needs. During the real-time receiving process, the order of data packets should be reorganized and errors should be corrected to ensure data integrity. Data verification mechanisms, such as CRC or SHA, can effectively detect errors and packet loss in data transmission, and promptly re-request the transmission of erroneous data packets to improve the integrity and reliability of data reception.
[0096] After receiving the complete video data packet, use the corresponding decoding tool or library to decode the H.265 video stream. Software decoders such as FFmpeg and hardware decoders such as NVIDIANVDEC can be selected according to the system configuration. FFmpeg decoding commands such as ffmpeg-iinput.h265-c:v hevc output.mp4 can directly decode the video stream. Hardware decoding requires ensuring that the device and driver support hardware acceleration, configuring decoding parameters to match the settings used during encoding, and ensuring that the decoded image quality and frame rate are consistent with the original data. During the decoding process, monitor the performance and output results of the decoder in real time to detect and handle possible decoding errors or frame losses.
[0097] After successful decoding, the video frames are converted into a format suitable for display (such as RGB images) and presented to the user or saved as a specific image file format (such as BMP or PNG). The display device needs to be configured to support the decoded resolution and frame rate, and display the image through a dedicated software interface or medical monitoring system. In terms of storage, according to the needs of the medical system, the image frames are stored in a suitable database or file system, such as using the Dicom format to store medical images for later access and processing. During the display and storage process, image quality verification and storage optimization strategies need to be considered to ensure that the image is presented clearly and the data storage is efficient and reliable.
[0098] As a preferred embodiment, the receiving end 5 is further used to store the first data and / or the decoded first data;
[0099] The receiving end 5 includes a remote device and / or a cloud server.
[0100] Specifically, the receiving end uses a remote device such as a PC, tablet or smart phone, and can be selectively connected to a cloud server. The receiving end 5 can receive and decode H.265 data and restore it to an ultrasound image; it can visualize the decoded image and support doctors to perform remote diagnosis. At the same time, the receiving end supports storing the received data or decoded images locally or in a cloud server for subsequent analysis or backup.
[0101] As a data storage and processing center, the cloud server receives ultrasound image data uploaded from different devices and performs centralized storage and management. The cloud server supports high concurrent access and provides fast data access and sharing services for multiple remote devices. At the same time, the cloud server can be configured with more powerful computing resources for further image analysis and processing.
[0102] Furthermore, the system of the present invention can distribute processing tasks to multiple CPU cores through multithreading technology, such as using OpenMP or Pthread library to realize parallelization of data processing tasks. CUDA or OpenCL technology used by GPU acceleration can significantly improve the processing speed of encoding and decoding, especially when processing large-scale data. Configuring encoders and decoders to use GPU acceleration ensures that image processing tasks can make full use of hardware resources, greatly improves the processing efficiency and response speed of the overall system, reduces the load of the CPU, and realizes efficient data stream processing.
[0103] During system operation, various resource indicators (such as CPU usage, GPU usage, memory usage, and network bandwidth) are monitored in real time. System monitoring tools (such as Prometheus, Grafana, or nvidia-smi) can be used to obtain these resource usage in real time and identify bottlenecks and anomalies. Based on the monitoring data, task allocation and resource usage strategies are dynamically adjusted, such as increasing the thread pool size or adjusting task priorities during peak loads, and reducing resource usage to save energy when idle. Resource management strategies include dynamic voltage and frequency scaling (DVFS) and process scheduling optimization, so that the system can run efficiently and remain stable under different load conditions.
[0104] For systems involving data transmission, intelligent bandwidth management is used to reasonably allocate bandwidth resources, avoid congestion, and automatically adjust the video transmission quality when network conditions are poor (adaptive bit rate technology ABR). At the network transmission layer, high-quality transmission protocols such as QUIC are used, combined with congestion control and error repair mechanisms (such as FEC forward error correction) to effectively reduce transmission delays and improve transmission efficiency. Through real-time network status monitoring, rapid response to network fluctuations, and dynamic adjustment of transmission strategies, high-quality data transmission can be achieved in different network environments, ensuring the efficiency and stability of the overall operation of the system.
[0105] The present invention also provides an ultrasonic image processing method based on the H.265 video encoding format, which is applied to the ultrasonic image processing system based on the H.265 video encoding format as described above, and the method includes:
[0106] S100, preprocessing the received ultrasonic image data to obtain preprocessed ultrasonic image data; wherein the ultrasonic image data is stored in a first predetermined format;
[0107] S200, adjusting encoding parameters according to image features of the preprocessed ultrasound image data to encode and obtain first data having a second predetermined format; the second predetermined format is an H.265 video encoding format.
[0108] The present invention also provides a handheld ultrasound device, comprising the above-mentioned ultrasound image processing system.
[0109] The advantages or beneficial effects of adopting the above technical solution are: the present invention adjusts the encoding parameters of H.265 video encoding according to the image characteristics of the ultrasound image data, thereby achieving efficient compression of the ultrasound image data while avoiding the loss of ultrasound information and the generation of artifacts, thereby improving the quality of the ultrasound image and providing a more accurate and reliable image basis for medical diagnosis.
[0110] The above are only preferred embodiments of the present invention, and are not intended to limit the implementation methods and protection scope of the present invention. Those skilled in the art should be aware that all solutions obtained by equivalent substitutions and obvious changes made using the contents of this specification and illustrations should be included in the protection scope of the present invention.
Claims
1. An ultrasonic image processing system based on H.265 video encoding format, characterized in that: include: An ultrasonic data processing module, used for preprocessing the received ultrasonic image data to obtain preprocessed ultrasonic image data; wherein the ultrasonic image data is stored in a first predetermined format; An image encoding module is connected to the ultrasound data processing module and is used to adjust encoding parameters according to image features of the preprocessed ultrasound image data to encode first data having a second predetermined format; the second predetermined format is an H.265 video encoding format.
2. The ultrasonic image processing system according to claim 1, characterized in that: The ultrasonic data processing module comprises: a data acquisition unit, configured to acquire ultrasonic signals received by the ultrasonic probe, digitally process the ultrasonic signals, and generate the ultrasonic image data stored in the first predetermined format; The image preprocessing unit is connected to the data acquisition unit and is used to preprocess the ultrasound image data to obtain the preprocessed ultrasound image data.
3. The ultrasonic image processing system according to claim 1, characterized in that: The first predetermined format is a Digital Imaging and Communications in Medicine format.
4. The ultrasonic image processing system according to claim 1, characterized in that: The image encoding module comprises: An analysis unit, configured to perform image analysis on the preprocessed ultrasound image data to obtain an image analysis result; the image analysis result includes feature information of different feature types; an image division unit, connected to the analysis unit, and configured to divide the preprocessed ultrasound image data into a plurality of key areas according to feature information in the image analysis result; A parameter adjustment unit, connected to the image division unit, configured to adjust the encoding parameters of the encoding unit according to the feature type of each key area; each key area corresponds to one encoding unit; An image encoding unit is connected to the parameter adjustment unit and is used to perform image encoding on the preprocessed ultrasound image data according to the adjusted encoding parameters to obtain the first data.
5. The ultrasonic image processing system according to claim 4, characterized in that: The parameter adjustment unit is used to adjust the encoding parameters according to the feature type of each key area based on the corresponding relationship between the feature type and the encoding parameter.
6. The ultrasonic image processing system according to claim 1, characterized in that: The feature types of the image features include image noise features and / or image texture features; The coding parameters include quantization parameters and prediction coding parameters. The prediction coding parameters include intra-frame prediction mode and inter-frame prediction mode. The intra-frame prediction mode includes a plane mode, a direct current mode and an angle mode.
7. The ultrasonic image processing system according to claim 1, characterized in that: Also includes: A quality assessment module, connected to the image encoding module, configured to perform quality assessment on the first data to obtain a quality assessment result; The image encoding module is further configured to readjust the encoding parameters when the quality assessment result indicates that the quality does not meet the standard.
8. The ultrasonic image processing system according to claim 1, characterized in that: Also includes: A data transmission module, connected to the image encoding module, and configured to transmit the first data having the second predetermined format to a receiving end via wireless transmission; The receiving end is used to decode the first data and visually display the decoded first data.
9. The ultrasonic image processing system according to claim 8, characterized in that: The receiving end is further used to store the first data and / or the decoded first data; The receiving end includes a remote device and / or a cloud server.
10. A method for processing ultrasonic images based on H.265 video encoding format, characterized in that: Applied to the ultrasonic image processing system according to any one of claims 1 to 9, the method comprises: Preprocessing the received ultrasonic image data to obtain preprocessed ultrasonic image data; wherein the ultrasonic image data is stored in a first predetermined format; The encoding parameters are adjusted according to the image features of the preprocessed ultrasound image data to encode the first data in a second predetermined format; the second predetermined format is the H.265 video encoding format.
Citation Information
Cited By
Low-delay video security exchange method suitable for video monitoring center
CN121056640A
Low-latency video security exchange method for video surveillance center
CN121056640B
