Image processing apparatus and imaging apparatus
By identifying flowers in the image processing device and performing HDR synthesis, the problem of flower image color saturation and insufficient overall brightness is solved, and high-quality output of the image is achieved.
Patent Information
- Application Number
- JP2023185133
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-10-30
- Publication Date
- 2025-05-14
AI Technical Summary
Flower images cause color saturation due to high brightness, reducing exposure to prevent saturation from causing overall image to be too dark.
By introducing units and decision units for identifying flowers in the image processing device, high dynamic range (HDR) synthesis is performed when flowers are identified, and the exposure of multiple images is adjusted by automatic exposure to synthesize HDR images with appropriate brightness.
It effectively prevents the color saturation of the flower image, while maintaining the overall brightness of the image, ensuring the quality of the image.
Smart Images

Figure 2025074381000001_ABST
Abstract
Description
[Technical field]
[0001] The present disclosure relates to an image processing device and an imaging device. [Background technology]
[0002] Patent Document 1 discloses an information processing device. In the information processing device, learning data and a test image are acquired. The learning data is used to perform learning for image processing such as SDR-HDR conversion, and an image conversion model is generated. The image conversion model is used to generate an inference image from the test image. The inference image is evaluated and an evaluation value of the inference image is calculated. Whether or not to perform re-learning is determined based on the evaluation value. The material of the subject appearing in the test image is recognized. The recognized materials include flowers. Depending on the combination of the recognized materials, objects appearing in the test image are recognized. Based on the recognized objects, the shooting environment of the captured image used for re-learning is determined (paragraphs 0013, 0060, 0062, 0064, 0069, 0072, 0077, 0082, 0084, 0085, and 0118). [Prior art documents] [Patent documents]
[0003] [Patent Document 1] International Publication No. 2022 / 130846 Summary of the Invention [Problem to be solved by the invention]
[0004] The image of the flower contained in the image has a high brightness. Therefore, the color of the flower image is easily saturated. Therefore, in order to prevent the color of the flower image from becoming saturated, it is effective to lower the exposure when the flower is captured. However, if the exposure is lowered so as to prevent the color from becoming saturated, the entire image becomes too dark.
[0005] In view of this problem, an aspect of the present disclosure provides an image processing device and an imaging device that can, for example, ensure that the brightness of the entire image is appropriate while preventing color saturation in the image of a flower included in the image. [Means for solving the problem]
[0006] An image processing device according to a first aspect of the present disclosure includes a recognition unit that recognizes an image of a flower contained in an image, and a determination unit that determines to perform high dynamic range synthesis when the image is recognized by the recognition unit.
[0007] An imaging device of a second aspect of the present disclosure includes an image processing device of the first aspect of the present disclosure, an imaging unit that performs imaging, and an automatic exposure unit that adjusts exposure in the imaging and causes the imaging unit to acquire multiple images that are synthesized into high dynamic range images. [Brief description of the drawings]
[0008] [Figure 1] 1 is a block diagram of an imaging device according to a first embodiment. [Diagram 2] 4 is a flowchart showing the flow of processing performed by the imaging device of the first embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0009] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. In the drawings, the same or equivalent elements are denoted by the same reference numerals, and duplicated descriptions will be omitted.
[0010] 1. First embodiment 1.1 Imaging device FIG. 1 is a block diagram of an imaging device according to the first embodiment.
[0011] The imaging device 1 shown in FIG. 1 captures an image to obtain it.
[0012] As shown in FIG. 1, the imaging device 1 includes an imaging section 11, an automatic exposure section 12, and an image processing device 13.
[0013] The imaging unit 11 captures an image to obtain an image 31. The imaging unit 11 includes an image sensor.
[0014] The automatic exposure unit 12 adjusts the exposure in imaging so that an image 31 having an appropriate brightness can be obtained. When high dynamic range (HDR) synthesis is performed by the image processing device 13, the automatic exposure unit 12 adjusts the exposure in imaging to cause the imaging unit 11 to obtain a plurality of images 32 having different exposures required for HDR synthesis. The plurality of images 32 is composed of three images, a ±0 image 41 having an appropriate brightness, a + image 42 having a brightness brighter than the appropriate brightness, and a - image 43 having a brightness darker than the appropriate brightness. The plurality of images 32 may be composed of two images, or may be composed of four or more images.
[0015] The image processing device 13 processes the acquired image 31. When the image 31 has a large difference in brightness, the image processing device 13 causes the automatic exposure unit 12 to adjust the exposure in the image capture so that multiple images 32 are acquired, and performs HDR synthesis of the acquired multiple images 32 to acquire an HDR image. Even when the image 31 has a small difference in brightness, when the image processing device 13 recognizes an image of a flower included in the image 31, the image processing device 13 causes the automatic exposure unit 12 to adjust the exposure in the image capture so that multiple images 32 are acquired, and performs HDR synthesis of the acquired multiple images 32 to acquire an HDR image. The acquired HDR image has a wide dynamic range.
[0016] 1.2 Image Processing Device As shown in FIG. 1, the image processing device 13 includes a recognition unit 21, a determination unit 22, and a synthesis unit .
[0017] The recognition unit 21 recognizes the image of a flower included in the image 31. The recognition unit 21 recognizes the image of a flower included in the image 31, for example, by artificial intelligence (AI) image analysis.
[0018] The determination unit 22 performs a primary determination and a secondary determination. In the primary determination, the determination unit 22 determines whether or not to perform HDR combining based on the contrast of the image 31. In the secondary determination, the determination unit 22 determines whether or not to perform HDR combining based on the recognition result 33 of the recognition unit 21.
[0019] The determination unit 22 determines to perform HDR combining when the brightness difference of the image 31 is equal to or greater than a reference level in the primary determination. The determination unit 22 determines not to perform HDR combining when the brightness difference of the image 31 is not equal to or greater than a reference level in the primary determination. For example, the determination unit 22 determines to perform HDR combining when the primary determination satisfies a first condition that the image 31 contains more very bright areas than a set amount and a second condition that the image 31 contains more very dark areas than a set amount. The determination unit 22 determines not to perform HDR combining when the primary determination does not satisfy the first condition or the second condition.
[0020] The determination unit 22 performs a secondary determination when it is determined in the primary determination that HDR compositing is not to be performed. In the secondary determination, the determination unit 22 determines that HDR compositing is to be performed when the recognition unit 21 recognizes an image of a flower included in the image 31. In the secondary determination, the determination unit 22 determines that HDR compositing is not to be performed when the recognition unit 21 does not recognize an image of a flower included in the image 31.
[0021] The synthesis unit 23 performs HDR synthesis based on the determination result 34 of the determination unit 22. When it is determined that HDR synthesis is to be performed, the synthesis unit 23 performs HDR synthesis of a plurality of images 32 to obtain an HDR image.
[0022] The acquired HDR image may be converted into a compressed image such as a JPEG image.
[0023] The image processing device 13 includes a processor, a memory, and peripheral circuits. The processor executes an image processing program stored in the memory, and causes the processor, the memory, and the peripheral circuits to operate as a recognition unit 21, a determination unit 22, and a synthesis unit 23. A part or all of the processing performed by the processor may be performed by a dedicated circuit.
[0024] 1.3 Processing flow FIG. 2 is a flowchart showing the flow of processing performed by the imaging device of the first embodiment.
[0025] The imaging device 1 executes steps S101 to S113 shown in FIG.
[0026] In step S101, the imaging unit 11 captures an image and obtains the image 31.
[0027] In the next step S102, the determination unit 22 determines whether or not the contrast of the image 31 is equal to or greater than the reference level. If the determination unit 22 determines that the contrast of the image 31 is equal to or greater than the reference level, steps S103 to S106 are executed. If the determination unit 22 determines that the contrast of the image 31 is not equal to or greater than the reference level, step S107 is executed.
[0028] In step S103, the determination unit 22 determines to perform HDR combining.
[0029] In step S104, the automatic exposure unit 12 sets the exposure difference between the multiple images 32 to a first exposure difference.
[0030] In step S105, the imaging unit 11 performs exposure bracketing photographing to obtain a first plurality of images having a first exposure difference.
[0031] In step S106, the synthesis unit 23 performs HDR synthesis on the first plurality of images.
[0032] Through steps S103 to S106, when the determination unit 22 determines in the primary determination that HDR combination is to be performed, the combination unit 23 performs HDR combination of the first plurality of images having a first exposure difference.
[0033] In step S107, the determination unit 22 determines whether or not the recognition unit 21 has recognized an image of a flower included in the image 31. If the recognition unit 21 has recognized an image of a flower included in the image 31, steps S108 to S111 are executed. If the recognition unit 21 has not recognized an image of a flower included in the image 31, steps S112 and S113 are executed.
[0034] In step S108, the determination unit 22 determines to perform HDR combining.
[0035] In step S109, the automatic exposure unit 12 sets the exposure difference between the multiple images 32 to the second exposure difference.
[0036] In step S110, the imaging unit 11 performs exposure bracketing photographing to obtain a second plurality of images having a second exposure difference.
[0037] In step S111, the synthesis unit 23 performs HDR synthesis on the second plurality of images. This makes it possible to ensure that the entire image 31 has appropriate brightness while preventing color saturation in the image of the flower contained in the image 31.
[0038] Through steps S108 to S111, when the determination unit 22 determines in the secondary determination that HDR combination is to be performed, the combination unit 23 performs HDR combination of the first plurality of images having the second exposure difference.
[0039] The second exposure difference is smaller than the first exposure difference because the exposure difference suitable for HDR synthesis performed when the image 31 includes an image of a flower is smaller than the exposure difference suitable for HDR synthesis performed when the image 31 has a large contrast between light and dark.
[0040] In step S112, the determination unit 22 determines not to perform HDR combining.
[0041] In step S113, the imaging unit 11 acquires an image.
[0042] The present disclosure is not limited to the above-described embodiments, and may be replaced with a configuration that is substantially the same as the configuration shown in the above-described embodiments, a configuration that has the same action and effect, or a configuration that can achieve the same purpose. [Explanation of symbols]
[0043] 1 imaging device, 11 imaging section, 12 automatic exposure section, 13 image processing device, 21 recognition section, 22 judgment section, 23 synthesis section, 33 recognition result, 34 judgment result, 31 image, 32 multiple images, 41 ±0 image, 42 + image, 43 - image.
Claims
1. a recognition unit that recognizes an image of a flower included in an image; a determination unit that determines to perform high dynamic range composition when the image is recognized by the recognition unit; An image processing device comprising:
2. The determination unit performs a primary determination to determine whether or not to perform the high dynamic range compositing based on a brightness difference of the image, and determines to perform the high dynamic range compositing when the primary determination determines that the high dynamic range compositing is not to be performed but the image is recognized by the recognition unit. The image processing device according to claim 1 .
3. a synthesis unit that performs high dynamic range synthesis of a first plurality of images having a first exposure difference when the determination unit determines in the primary determination that the high dynamic range synthesis is to be performed, and that performs high dynamic range synthesis of a second plurality of images having a second exposure difference smaller than the first exposure difference when the image is recognized by the recognition unit and it is determined that the high dynamic range synthesis is to be performed. The image processing device according to claim 2 .
4. An image processing device according to any one of claims 1 to 3, An imaging unit that captures an image; an automatic exposure unit that adjusts exposure during the imaging to cause the imaging unit to acquire a plurality of images that are to be synthesized into a high dynamic range; An imaging device comprising:
Citation Information
Patent Citations
Information processing device, information processing system, information processing method, and program
WO2022130846A1