Endoscope image adaptive processing system
By setting the controller and memory in the endoscopic image processing system to store the image sensor ID and storing multiple ID collections on the host, using the FPGA module to match the configuration parameters, adaptive processing of different image sensors is realized, and the current solution is to solve the problem of There are cumbersome issues in the technology that firmware updates are complicated, which improves processing efficiency.
Patent Information
- Application Number
- CN202420021473.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- Filing Date
- 2024-01-04
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2034-01-04
AI Technical Summary
When replacing different types of image sensors, existing endoscopic image processing systems need to update the firmware, which is cumbersome and inefficient, so they cannot adapt to different image sensors for processing.
Design an endoscopic image adaptive processing system, by setting the controller and memory in the endoscope to store the ID of the current image sensor, and setting the FPGA module and memory on the host to store the ID collection of multiple image sensors, the FPGA module is According to the ID matching initialization configuration parameters, it adapts to different image sensors for processing without updating the firmware.
It simplifies the operation process, improves the efficiency of endoscopic image processing, and can adapt to different image sensors for processing, reducing the complexity when replacing sensors.
Smart Images

Figure CN221949763U_ABST
Abstract
Description
Technical Field
[0001] The utility model belongs to the technical field of image processing, and in particular relates to an endoscope image adaptive processing system. Background Art
[0002] Endoscopes and image processors are used together to detect lesions in patients. Endoscopes are usually equipped with micro-image sensors, such as CMOS and CCD, to collect image information of patient lesions, and transmit the image information to the image processor after preprocessing. The image processor obtains the front-end image information, converts it, and outputs it to the display screen for display. In order to adapt to different scenarios, it is necessary to replace different types of endoscopes, such as nasopharyngeal endoscopes, gastrointestinal endoscopes, or renal pelvis endoscopes, etc. The image sensors configured in these different types of endoscopes may be different. In the prior art, the same host (i.e., image processor) is usually used to process the images of different endoscopic image sensors. In order to adapt to the output when different image sensors are replaced, the firmware is often updated to adapt to different image sensors. The operation is more cumbersome, resulting in low efficiency in endoscopic image processing. Summary of the invention
[0003] The utility model aims to provide an endoscope image adaptive processing system, which can adaptively process images using different image sensors without updating firmware, thereby simplifying the operation and improving the endoscope image processing efficiency.
[0004] To achieve the above-mentioned purpose, the utility model provides an endoscope image adaptive processing system, comprising a host and multiple endoscopes, wherein each of the endoscopes is adapted to have an image sensor of different models, each of the endoscopes is provided with a controller and a first memory, the controller is communicatively connected to the first memory, the host is provided with an FPGA module and a second memory, the image sensor is communicatively connected to the FPGA module, the controller is communicatively connected to the FPGA module, and the FPGA module is communicatively connected to the second memory.
[0005] In some embodiments, the controller is communicatively connected to the FPGA module via a UART protocol.
[0006] In some embodiments, the image sensor is communicatively connected to the FPGA module via an I2C protocol.
[0007] In some embodiments, the controller is communicatively connected to the first memory via an I2C protocol.
[0008] In some embodiments, the FPGA module is provided with an input port, an image processor, and an output port connected in sequence, and the image sensor is communicatively connected to the input port.
[0009] In some embodiments, the endoscope is further provided with a coding serialization module, and the host is further provided with a serialization decoding module. The coding serialization module and the serialization decoding module are connected via a transmission line, the image sensor is communicatively connected to the coding serialization module, and the serialization decoding module is communicatively connected to the input port.
[0010] In some embodiments, the input port is configured as a MIPI CSI-RX interface.
[0011] In some embodiments, the FPGA module is further provided with a REST interface, and the REST interface is connected to the MIPICSI-RX interface.
[0012] In some embodiments, the output port is configured as a VO channel interface.
[0013] In some embodiments, the host is further provided with a high-definition multimedia interface, and the high-definition multimedia interface is connected to the VO channel interface.
[0014] The beneficial effect of the utility model lies in that, by setting a first memory in the endoscope for storing the identity document (ID) of the current image sensor, and setting an FPGA module and a second memory on the host, the second memory is used to store the ID collection of multiple image sensors, the FPGA module can match the ID collection according to different image sensor IDs to identify the current image sensor, and determine the corresponding initialization configuration parameters, so as to enter different working states, so that image processing can be adaptively performed on different image sensors without updating the firmware, thereby simplifying the operation and improving the efficiency of endoscope image processing. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 The present invention is a structural schematic diagram of an endoscope image adaptive processing system.
[0016] Description of reference numerals:
[0017] 10. Endoscope; 20. Host. DETAILED DESCRIPTION
[0018] In order to facilitate the understanding of the present invention, specific embodiments of the present invention will be described in more detail below with reference to the accompanying drawings.
[0019] Unless otherwise specified or defined, the "first, second..." used in this article is merely used to distinguish names and does not represent a specific quantity or order.
[0020] Unless specifically stated or defined otherwise, the term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.
[0021] It should be noted that, in this document, “fixed to” or “connected to” may mean directly fixing or connecting to an element, or indirectly fixing or connecting to an element.
[0022] like Figure 1 As shown, the embodiment of the utility model discloses an endoscope image adaptive processing system, including multiple endoscopes 10 and a host 20, wherein each endoscope 10 is adapted to have image sensors of different models, each endoscope 10 is provided with a controller (Microcontroller Unit, MCU) and a first memory, the MCU is communicatively connected to the first memory, the host 20 is provided with an FPGA module and a second memory, the image sensor is communicatively connected to the FPGA module, the MCU is communicatively connected to the FPGA module, the FPGA module is communicatively connected to the second memory, the first memory is used to store the current image sensor ID configured by the endoscope 10, and the second memory is used to store a collection of IDs of image sensors of different models. The endoscope can match image sensors of different models to adapt to different clinical scenarios and meet different medical image requirements. In actual application, the image sensor is integrated in the endoscope, and the endoscope needs to be replaced each time a different type of image sensor is replaced. Figure 1 The middle dotted line indicates that the endoscope 10 and the host 20 are connected via an interface socket and can be plugged in and out.
[0023] In this embodiment, a first memory is set on the endoscope 10 for storing the ID of the current image sensor, and an FPGA module and a second memory are set on the host 20, and the second memory is used to store the ID collection of multiple image sensors. The FPGA module can match the ID collection according to different image sensor IDs to identify the current image sensor, and determine the corresponding initialization configuration parameters to enter different working states. Therefore, it can adaptively process images for different image sensors without updating the firmware, thereby simplifying the operation and improving the efficiency of endoscope image processing.
[0024] Among them, the FPGA module is set to its working state by the program stored in the on-chip RAM. The on-chip RAM is usually a memory embedded in the microprocessor or microcontroller chip. Therefore, the on-chip RAM of the FPGA module needs to be programmed during operation. Users can use different programming methods according to different configuration modes. When powered on, the FPGA module reads the stored configuration data into the on-chip RAM. After the configuration is completed, the FPGA module enters the working state. After power failure, the FPGA module returns to a blank chip, and the internal logical relationship disappears. Therefore, the FPGA module can be used repeatedly. The programming of the FPGA module does not require a dedicated FPGA programmer, but only a general EPROM or PROM programmer. When the FPGA function needs to be modified, just replace an EPROM. In this way, the same FPGA module and different programming data can produce different circuit functions. Therefore, the use of the FPGA module is very flexible.
[0025] In this embodiment, the MCU in the endoscope 10 is connected to the FPGA module through the UART protocol, and the different image sensors are connected to the FPGA module through the I2C protocol, and the MCU is connected to the first memory through the I2C protocol. Specifically, when working, the MCU reads the current image sensor ID fixed on the first memory through the I2C protocol and sends it to the FPGA module through the UART protocol. The FPGA module matches the received current image sensor ID with the collection of multiple image sensor IDs stored in the second memory. If the current image sensor ID is matched from the collection of image sensor IDs, the image data transmission can be executed. If the current image sensor ID cannot be matched, it means that the current image sensor is not in the library and cannot be identified. New ID information can be manually entered to allow the image sensor with this ID to establish a connection later.
[0026] Exemplarily, both the first memory and the second memory can be configured as an electrically erasable programmable read only memory (EEPROM), which is a storage chip that does not lose data after power failure, and can erase existing information and reprogram on a computer or dedicated device. It is generally used in plug-and-play chips.
[0027] In this embodiment, the FPGA module is provided with an input port, an image processor (Image Signal Processing, ISP) and an output port connected in sequence, and the image sensor is communicatively connected to the input port. Exemplarily, the input port is set to a MIPI CSI-RX interface, and the output port is set to a VO channel interface. Further, the host 20 is also provided with a High Definition Multimedia Interface (HDMI), and HDMI is connected to the VO channel interface. Among them, the image sensor can transmit the image data collected by it to the MIPI CSI-RX interface via MIPI, and the MIPI CSI-RX interface is used to receive the image data collected by the image sensor, and transmit it to the ISP for image processing. After the ISP processes the image data, it is transmitted to the VO channel interface, and then transmitted to the HDMI for output display. Among them, the host 20 can be connected to an electronic display screen via HDMI for image display.
[0028] Among them, the MIPI CSI-RX interface can support a variety of image sensors, including CMOS and CCD sensors, and can also support a variety of image formats, such as RAW, YUV, RGB, etc. At the same time, the MIPI CSI-RX interface adopts serial transmission, which can reduce power consumption and has high reliability, which can ensure the stability and security of image data transmission.
[0029] In some embodiments, the FPGA module is also provided with a REST interface and an IO interface, and the REST interface is connected to the MIPI CSI-RX interface. The IO interface is used to receive a reset instruction and transmit it to the REST interface. The REST interface is used to reset the MIPI CSI-RX interface after the FPGA module determines which image sensor it is by comparing the ID to ensure that the MIPI clock is not confused. The FPGA module is used to load the corresponding configuration parameters through the I2C protocol and drive the initialization image sensor after the MIPI CSI-RX interface is reset.
[0030] In order to increase the transmission distance of the endoscope signal, in this embodiment, the endoscope 10 is also provided with a coding serialization module, and the host 20 is also provided with a serialization decoding module. The coding serialization module and the serialization decoding module are connected through a transmission line, the image sensor is connected to the coding serialization module, and the serialization decoding module is connected to the input port (i.e., the MIPI CSI-RX interface). Among them, the image sensor is connected to the coding serialization module through MIPI, and the coding serialization module and the serialization decoding module are transmitted through the V-by-One protocol. After serialization, decoding is required to restore the standard MIPI signal and to connect with the back-end FPGA module chip. That is, the coding serialization module is used to convert the RAW10 data of MIPI 2Lane into the RAW10 data of V-by-One@HS1Lane, and then transmit it to the serialization decoding module through the transmission line. The serialization decoding module is used to convert the RAW10 data of V-by-One@HS1Lane back to the RAW10 data of MIPI 2Lane.
[0031] By setting up the encoding serialization module and the serialization decoding module, the transmission line can be extended. Compared with the current MIPI signal, the transmission distance after enhancement is limited to 3 meters. By using encoding serialization and decoding serialization to increase the signal transmission distance, the limitation of 3 meters in transmission length can be broken.
[0032] The above embodiments are not exhaustive enumerations based on the present invention, and there may be multiple other implementations not listed. Any replacement and improvement made without violating the concept of the present invention shall fall within the protection scope of the present invention.
Claims
1. An endoscopic image adaptive processing system, characterized in that: The invention comprises a host and a plurality of endoscopes, wherein each of the endoscopes is adapted to be equipped with an image sensor of a different model, each of the endoscopes is provided with a controller and a first memory, the controller is communicatively connected to the first memory, the host is provided with an FPGA module and a second memory, the image sensor is communicatively connected to the FPGA module, the controller is communicatively connected to the FPGA module, and the FPGA module is communicatively connected to the second memory.
2. The endoscopic image adaptive processing system according to claim 1, characterized in that: The controller is connected to the FPGA module for communication via the UART protocol.
3. The endoscopic image adaptive processing system according to claim 1, characterized in that: The image sensor is connected to the FPGA module through the I2C protocol.
4. The endoscopic image adaptive processing system according to claim 1, characterized in that: The controller is connected to the first memory through the I2C protocol.
5. The endoscopic image adaptive processing system according to any one of claims 1 to 4, characterized in that: The FPGA module is provided with an input port, an image processor and an output port which are connected in sequence, and the image sensor is communicatively connected with the input port.
6. The endoscopic image adaptive processing system according to claim 5, characterized in that: The endoscope is also provided with a coding serialization module, and the host is also provided with a serialization decoding module. The coding serialization module and the serialization decoding module are connected via a transmission line, the image sensor is communicatively connected to the coding serialization module, and the serialization decoding module is communicatively connected to the input port.
7. The endoscopic image adaptive processing system according to claim 5, characterized in that: The input port is set to a MIPI CSI-RX interface.
8. The endoscopic image adaptive processing system according to claim 7, characterized in that: The FPGA module is also provided with a REST interface, and the REST interface is connected to the MIPI CSI-RX interface.
9. The endoscopic image adaptive processing system according to claim 5, characterized in that: The output port is set as a VO channel interface.
10. The endoscopic image adaptive processing system according to claim 9, characterized in that: The host is also provided with a high-definition multimedia interface, and the high-definition multimedia interface is connected to the VO channel interface.
