Data source and test system and method for image target recognition

Abstract

The invention provides a data source for image target recognition and a test system method, the system comprises a foreground module, a background module, a synthesis module and an analysis module, the foreground module is used for receiving and storing a foreground image, the background module is used for receiving and storing a background image, and the synthesis module is used for analyzing the foreground image and the background image by using the received foreground image and the background image. The foreground module generates a test image and transmits the test image to the tested equipment, and output interfaces of the foreground module and the background module are both connected with the synthesis module; and the analysis module is used for evaluating and optimizing a target recognition algorithm by utilizing the received target generation information and the processing output of the tested equipment, analyzing the image recognition performance of the tested equipment and transmitting an obtained analysis result to the human-computer interface module. According to the invention, the large-data-volume image data source for target identification and the test system and method matched with the large-data-volume image data source are realized.

Description

Technical Field

[0001] The present invention relates to the field of image processing and target recognition, and in particular to a data source and a test system and method for image target recognition. Background Art

[0002] In today's digital and information age, image target recognition technology is playing an increasingly important role. It is widely used in many fields, such as security monitoring, autonomous driving, medical diagnosis, industrial detection, etc.

[0003] In image object recognition testing, obtaining sufficient test data is the key to evaluating algorithm performance and improving model accuracy. However, traditional methods often use the method of storing a large number of real scene images, which not only takes up a lot of storage space, but also limits the diversity and flexibility of data. In addition, each test needs to read the image from the storage device, resulting in unnecessary waste of time and resources.

[0004] In summary, a new data source, test system and method for image target recognition are needed to solve the problem of needing to store a large number of scene images and limited reading and writing speeds. Summary of the invention

[0005] In view of the defects in the prior art, the present invention aims to provide a large-scale image data source and an adaptive test system and method that can store foreground images and background images separately, utilize a small number of foreground images and background images, and generate target images under various backgrounds in real time through synthesis technology for target recognition.

[0006] A data source and test system for image target recognition provided by the present invention includes: a foreground module, a background module, a synthesis module and an analysis module;

[0007] The foreground module is used to receive and save the foreground image, and the input interface of the foreground module is connected to the external human-machine interface module;

[0008] The background module is used to receive and save the background image, and the input interface of the background module is connected to the human-machine interface module;

[0009] The synthesis module generates a test image using the received foreground image and background image, and transmits the test image to the device under test, and the output interfaces of the foreground module and the background module are both connected to the synthesis module;

[0010] The analysis module uses the target generation information from the human-machine interface module and the processing output of the device under test to evaluate and optimize the target recognition algorithm, analyze the image recognition performance of the device under test, and transmit the obtained analysis results to the human-machine interface module.

[0011] Preferably, the synthesis module adopts a GPU-based image synthesis method.

[0012] Preferably, the output interface of the synthesis module is connected to the device under test, the synthesis module synthesizes the received foreground image and background image to generate a synthesized image, and transmits the synthesized image to the device under test as a test image.

[0013] Preferably, the synthesis module further includes a simulation module and a selection module, the output interfaces of the synthesis module and the simulation module are both connected to the selection module, and the output interface of the selection module is connected to the device under test;

[0014] The simulation module is used to receive and save the simulation image, the input interface of the simulation module is connected to the human-machine interface module, and the human-machine interface module transmits the simulation image to the simulation module;

[0015] The synthesis module synthesizes the received foreground image and background image to generate a synthesized image, transmits the synthesized image and the simulation image to the selection module, and transmits the output of the selection module as a test image to the device under test.

[0016] Preferably, the selection module uses a control switch to select channels of the received synthetic image and simulation image, and then transmits them to the device under test.

[0017] Preferably, the analysis module is connected to an output interface of the device under test.

[0018] Preferably, the analysis module further comprises a decompression module, the decompression module is connected to the analysis module, and the decompression module is connected to the output interface of the device under test;

[0019] The decompression module is used to decompress the output of the device under test as raw data and transmit the obtained decompressed image to the analysis module.

[0020] Preferably, the system also includes a temperature measurement and control power supply module, which is used to control and maintain the operating status of the system, monitor the temperature of the equipment system and control the power switch. The equipment under test is connected to the temperature measurement and control power supply module, and the output interface of the temperature measurement and control power supply module is connected to the human-machine interface module.

[0021] Preferably, the system further comprises a storage module for saving the output of the device under test, saving the decompressed image, saving the analysis result, and saving the operation information and log of the system;

[0022] The output interface of the analysis module is connected to the storage module, the output interface of the device under test is connected to the storage module, the output interface of the decompression module is connected to the storage module, the output interface of the temperature measurement and control power supply module is connected to the storage module, and the output interface of the storage module is connected to the human-machine interface module;

[0023] The analysis module transmits the analysis results to the storage module; the decompression module sends the original data and decompressed image to the storage module, and the temperature measurement and control power supply module sends the system operation information and log to the storage module.

[0024] The present invention also provides a data source and a testing method for image target recognition, comprising the following steps:

[0025] S1: obtaining a foreground image through a foreground module, obtaining a background image through a background module, generating a composite image from the foreground image and the background image according to set requirements, and selecting the composite image as an image input of the device under test;

[0026] S2: Acquire a simulation image through a simulation module, and after selection, use it as the image input of the device under test;

[0027] S3: obtaining the output of the device under test as raw data, decompressing it through a decompression module to obtain a decompressed image, and then analyzing it through an analysis module to obtain an analysis result, sending the analysis result to a human-machine interface module, and sending the raw data, the decompressed image and the analysis result to a storage module for storage;

[0028] S4: Obtain the measurement and control, temperature, and power supply of the measurement and control temperature power supply module, display and control through the human-machine interface module, and send the operation information and log to the storage module for storage.

[0029] Compared with the prior art, the present invention has the following beneficial effects:

[0030] The present invention stores the foreground image and the background image separately, and uses a small amount of foreground images and background images to generate target images under various backgrounds in real time through synthesis technology. This method not only saves storage space, but also can generate a large amount of diverse test data using limited images. At the same time, since the image can be dynamically synthesized, the actual scene in different environments can be simulated, thereby improving the authenticity and reliability of the test and meeting the demand for large data volume test image data set for target recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0032] Figure 1is a system framework diagram for image target recognition in an embodiment of the present invention;

[0033] Figure 2 The figure is a flow chart of a testing method for image target recognition in an embodiment of the present invention. DETAILED DESCRIPTION

[0034] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0035] The present invention stores foreground images and background images separately, uses a small amount of foreground images and background images, and generates target images under various backgrounds in real time through synthesis technology, thereby providing a large data source of images for target recognition and an adaptive test system and method.

[0036] Embodiment 1:

[0037] Figure 1 4 is a system framework diagram for image target recognition in an embodiment of the present invention.

[0038] like Figure 1 As shown, this embodiment provides a data source and test system for image target recognition, including a foreground module, a background module, a synthesis module, an analysis module, a simulation module, a selection module, a decompression module, a temperature measurement and control power supply module, and a storage module.

[0039] The foreground module is used to receive and save foreground images of various common targets. The input interface of the foreground module is connected to the external human-machine interface module.

[0040] The human-machine interface module transmits the foreground image to the foreground module.

[0041] The background module is used to receive and save background images, namely, images of various commonly used backgrounds. The input interface of the background module is connected to the human-machine interface module.

[0042] The human-machine interface module transmits the background image to the background module.

[0043] The synthesis module generates a test image using the received foreground image and background image, and transmits the test image to the device under test. The output interfaces of the foreground module and the background module are both connected to the synthesis module.

[0044] In this embodiment, the device under test is an intelligent processor.

[0045] The synthesis module adopts a GPU-based image synthesis method. Through a small number of foreground images and background images, it uses image synthesis to generate multiple target images under different backgrounds in real time, realizing real-time synthesis and real-time testing.

[0046] In one embodiment, the output interface of the synthesis module is connected to the device under test, the synthesis module synthesizes the received foreground image and background image to generate a synthesized image, and transmits the synthesized image to the device under test as a test image.

[0047] In this embodiment, the synthesis module further includes a simulation module and a selection module. The output interfaces of the synthesis module and the simulation module are both connected to the selection module, and the output interface of the selection module is connected to the device under test.

[0048] The simulation module is used to receive and save simulation images, that is, multiple target images under different backgrounds. The input interface of the simulation module is connected to the human-machine interface module, and the human-machine interface module transmits the simulation images to the simulation module.

[0049] The synthesis module synthesizes the received foreground image and background image to generate a synthesized image, transmits the synthesized image and the simulation image to the selection module, and transmits the output of the selection module as a test image to the device under test.

[0050] The selection module uses the control switch to select the channels of the received synthetic image and simulation image, and then transmits them to the device under test.

[0051] The analysis module uses the target generation information from the human-machine interface module and the processing output of the device under test to evaluate and optimize the target recognition algorithm, analyze the image recognition performance of the device under test, and transmit the obtained analysis results to the human-machine interface module.

[0052] The output interface of the analysis module is connected to the human-machine interface module, and the human-machine interface module transmits the target generation information to the analysis module.

[0053] The analysis results include information such as the accuracy of target recognition.

[0054] In one embodiment, the analysis module is connected to an output interface of the device under test.

[0055] In this embodiment, the analysis module further includes a decompression module, the decompression module is connected to the analysis module, and the decompression module is connected to the output interface of the device under test.

[0056] The decompression module is used to decompress the output of the device under test as raw data and transmit the obtained decompressed image to the analysis module.

[0057] The temperature measurement and control power supply module is used to control and maintain the operating status of the system, monitor the temperature of the equipment system and control the switch of the power supply. The equipment under test is connected to the temperature measurement and control power supply module, and the output interface of the temperature measurement and control power supply module is connected to the human-machine interface module.

[0058] The storage module is used to save the output of the device under test, save the decompressed image, save the analysis results, and save the system's operation information and logs.

[0059] The output interface of the analysis module is connected to the storage module, the output interface of the device under test is connected to the storage module, the output interface of the decompression module is connected to the storage module, the output interface of the temperature measurement and control power supply module is connected to the storage module, and the output interface of the storage module is connected to the human-machine interface module.

[0060] The analysis module transmits the analysis results to the storage module; the decompression module sends the original data and decompressed image to the storage module, and the temperature measurement and control power supply module sends the system operation information and log to the storage module.

[0061] In this embodiment, the storage module adopts a solid state hard disk, which effectively improves the reading and writing speed of data by increasing the storage space of the solid state hard disk.

[0062] Embodiment 2:

[0063] Figure 2 The figure is a flow chart of a testing method for image target recognition in an embodiment of the present invention.

[0064] like Figure 2 As shown, the embodiment of the present invention provides a data source and a test method for image target recognition, using the data source and the test system for image target recognition described in the above embodiment 1, including the following steps:

[0065] S1: obtaining a foreground image through a foreground module, obtaining a background image through a background module, generating a composite image from the foreground image and the background image according to set requirements, and selecting the composite image as an image input of the device under test;

[0066] S2: Acquire a simulation image through a simulation module, and after selection, use it as the image input of the device under test;

[0067] S3: obtaining the output of the device under test as raw data, decompressing it through a decompression module to obtain a decompressed image, and then analyzing it through an analysis module to obtain an analysis result, sending the analysis result to a human-machine interface module, and sending the raw data, the decompressed image and the analysis result to a storage module for storage;

[0068] S4: Obtain the measurement and control, temperature, and power supply of the measurement and control temperature power supply module, display and control through the human-machine interface module, and send the operation information and log to the storage module for storage.

[0069] Those skilled in the art know that, in addition to realizing the system and its various devices, modules, and units provided by the present invention in a purely computer-readable program code, it is entirely possible to realize the same functions in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system and its various devices, modules, and units provided by the present invention can be considered as a hardware component, and the devices, modules, and units included therein for realizing various functions can also be regarded as structures within the hardware component; the devices, modules, and units for realizing various functions can also be regarded as both software modules for realizing the method and structures within the hardware component.

[0070] In the description of the present application, it should be understood that the terms "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0071] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A data source and test system for image target recognition, characterized in that: include: Foreground module, background module, synthesis module and analysis module; The foreground module is used to receive and save the foreground image, and the input interface of the foreground module is connected to the external human-machine interface module; The background module is used to receive and save the background image, and the input interface of the background module is connected to the human-machine interface module; The synthesis module generates a test image using the received foreground image and background image, and transmits the test image to the device under test, and the output interfaces of the foreground module and the background module are both connected to the synthesis module; The analysis module uses the target generation information from the human-machine interface module and the processing output of the device under test to evaluate and optimize the target recognition algorithm, analyze the image recognition performance of the device under test, and transmit the obtained analysis results to the human-machine interface module.

2. The data source and test system for image target recognition according to claim 1, characterized in that: The synthesis module adopts a GPU-based image synthesis method.

3. The data source and test system for image target recognition according to claim 1, characterized in that: The output interface of the synthesis module is connected to the device under test. The synthesis module synthesizes the received foreground image and the background image to generate a synthesized image, and transmits the synthesized image to the device under test as a test image.

4. The data source and test system for image target recognition according to claim 1, characterized in that: The synthesis module further comprises a simulation module and a selection module, the output interfaces of the synthesis module and the simulation module are both connected to the selection module, and the output interface of the selection module is connected to the device under test; The simulation module is used to receive and save the simulation image, the input interface of the simulation module is connected to the human-machine interface module, and the human-machine interface module transmits the simulation image to the simulation module; The synthesis module synthesizes the received foreground image and the background image to generate a synthesized image, transmits the synthesized image and the simulation image to the selection module, and transmits the output of the selection module as a test image to the device under test.

5. The data source and test system for image target recognition according to claim 4, characterized in that: The selection module uses a control switch to select channels for the received synthetic image and simulation image, and then transmits them to the device under test.

6. The data source and test system for image target recognition according to claim 1, characterized in that: The analysis module is connected to the output interface of the device under test.

7. The data source and test system for image target recognition according to claim 1, characterized in that: The analysis module further comprises a decompression module, the decompression module is connected to the analysis module, and the decompression module is connected to the output interface of the device under test; The decompression module is used to decompress the output of the device under test as raw data, and transmit the obtained decompressed image to the analysis module.

8. The data source and test system for image target recognition according to claim 7, characterized in that: The system also includes a temperature measurement and control power supply module, which is used to control and maintain the operating status of the system, monitor the temperature of the equipment system and control the switch of the power supply. The equipment under test is connected to the temperature measurement and control power supply module, and the output interface of the temperature measurement and control power supply module is connected to the human-machine interface module.

9. The data source and test system for image target recognition according to claim 8, characterized in that: The system further comprises a storage module for storing the output of the device under test, storing the decompressed image, storing the analysis result, and storing the operation information and log of the system; The output interface of the analysis module is connected to the storage module, the output interface of the device under test is connected to the storage module, the output interface of the decompression module is connected to the storage module, the output interface of the temperature measurement and control power supply module is connected to the storage module, and the output interface of the storage module is connected to the human-machine interface module; The analysis module transmits the analysis result to the storage module; the decompression module sends the original data and the decompressed image to the storage module, and the temperature measurement and control power supply module sends the system operation information and log to the storage module.

10. A data source and test method for image target recognition, using the data source and test system for image target recognition according to claim 9, characterized in that: The steps include: S1: obtaining a foreground image through a foreground module, obtaining a background image through a background module, generating a composite image from the foreground image and the background image according to set requirements, and selecting the composite image as an image input of the device under test; S2: Acquire a simulation image through a simulation module, and after selection, use it as the image input of the device under test; S3: obtaining the output of the device under test as raw data, decompressing it through a decompression module to obtain a decompressed image, and then analyzing it through an analysis module to obtain an analysis result, sending the analysis result to a human-machine interface module, and sending the raw data, the decompressed image and the analysis result to a storage module for storage; S4: Obtain the measurement and control, temperature, and power supply of the measurement and control temperature power supply module, display and control through the human-machine interface module, and send the operation information and log to the storage module for storage.

Images