Method and system for registering field of view of ultraviolet lens imaging and visible light lens imaging with combination lens and positioning support

Abstract

The application provides a kind of combination lens, positioning support, the field of view registration method and system of ultraviolet lens imaging and visible light lens imaging, including main visible light imaging lens, ultraviolet imaging lens, auxiliary registration imaging lens, ensure that auxiliary registration imaging lens and ultraviolet imaging lens are registered in view, under the premise that main visible light imaging lens and auxiliary registration imaging lens have overlap, collect the image data of main visible light imaging lens, auxiliary registration imaging lens, utilize DeepLabv3Plus to extract the high-level feature map of image data, calculate the relative translation, relative rotation angle of main visible light imaging lens and auxiliary registration imaging lens, and migrate as the corresponding information of main visible light imaging lens and ultraviolet imaging lens.The application adds micro lens as auxiliary registration imaging lens, realizes the registration of ultraviolet imaging field of view and visible light imaging field of view through indirect form, so that users can not only find ultraviolet signal, but also accurately locate signal source.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, specifically demonstrating a method and system for field registration of combined lenses, positioning supports, ultraviolet lens imaging, and visible light lens imaging. Background Technology

[0002] Ultraviolet (UV) imaging utilizes specialized instruments to receive the UV signals emitted by corona discharge. After special processing, these signals are overlaid with a visible light image. By combining these two images, the location and severity of the corona discharge can be clearly observed. The detection band of the UV detection instrument is typically selected within the solar blind zone. By detecting UV light at this wavelength, the intensity of the discharge in electrical equipment can be determined. "Solar blind zone" refers to a specific spectral segment within the UV light band (240nm–280nm). This band is unaffected by solar radiation, meaning it is insensitive to solar radiation. Within this band, almost all of the ultraviolet radiation from the sun is absorbed by the ozone layer, resulting in extremely low levels of UV light from the sun. Therefore, the detected UV light is emitted by the discharge of electrical equipment. Taking advantage of the fact that the UV light emitted by "solar blind" targets is stronger than solar UV radiation, a UV imager can observe and detect the UV target signal in this band. After converting the UV signal into a visible image signal for analysis and measurement, it is possible to determine whether a discharge exists in high-voltage transmission and transformation equipment. A comparison of the UV and visible light spectra of corona discharge is shown below. Figure 1 As shown.

[0003] An ultraviolet imager consists of a lens system, filters, a light sensor, a control and analysis system, and an image display system. The detection principle of an ultraviolet imager is as follows: Figure 2 As shown. The visible light lens captures a visible light image of the corona target, while a filter blocks light waves outside the "solar blind" region. The ultraviolet lens performs ultraviolet imaging on the filtered light waves. A light sensor performs photoelectric conversion on the corona image, and a control and analysis system analyzes the light intensity to control the entire detection system. The image display system fuses and displays these two types of image information, analyzing the mixed imaging information to determine the specific discharge status of the target.

[0004] Ultraviolet detection technology has the following characteristics:

[0005] (1) When testing power equipment, it is not necessary to shut down the power and stop other equipment from operating. It can also restore the discharge situation of power equipment under real operating conditions.

[0006] (2) Ultraviolet discharge detection does not require climbing to heights, and can detect electrical equipment from a distance, ensuring the personal safety of the testing personnel;

[0007] (3) Collect the basic data such as images and pictures generated by ultraviolet detection to form a basis for judging the strength of discharge, which will facilitate the state management and condition maintenance of the power system and provide a reliable theoretical basis for future periodic maintenance.

[0008] There are two common methods for fusing ultraviolet and visible light images: Method 1 is to deflect the visible light by 90 degrees using a beam splitter, while the ultraviolet light can pass through the beam splitter without changing its direction. In this case, the visible light camera and the ultraviolet light camera are arranged at a 90-degree angle to achieve the display effect of the optical axes of the visible light and ultraviolet light coinciding; Method 2 is to arrange the visible light camera and the ultraviolet light camera in parallel.

[0009] The advantage of Method 1 is that it can achieve high-precision fusion and superposition effects at any observation distance. The disadvantage is that the layout of the beam splitter and the camera after the optical path deflection will occupy a lot of space, ultimately making it impossible to miniaturize the product.

[0010] Method two offers the advantage of miniaturization, making it suitable for drones. However, due to the baseline distance between the center points of the two camera lenses, the center point of the visible light image is never the same as the center point of the ultraviolet image. Assuming their optical axes are nearly parallel, the closer the distance, the higher the pixel value of the offset between the center points of the two images in the photograph. This causes observational bias on the product side. Therefore, the common practice is to calibrate at frequently used observation distances and compensate for the offset through software. The problem with this approach is that a camera calibrated at 10 meters cannot always maintain a 10-meter shooting distance in actual use, and images taken at other distances will have deviations, resulting in inaccurate ultraviolet signal source localization. Summary of the Invention

[0011] In view of this, the purpose of the present invention is to provide a field-of-view registration method and registration system for a combination lens, a positioning support, ultraviolet lens imaging and visible light lens imaging, to solve the problem of inaccurate positioning of ultraviolet signal sources in existing ultraviolet detection applications.

[0012] One of the technical solutions is as follows:

[0013] A method for registering the fields of view of ultraviolet lens imaging and visible light lens imaging includes the following steps:

[0014] a) Select a primary visible light imaging lens (C1), select an ultraviolet imaging lens (UV1), and select a miniature lens as an auxiliary registration imaging lens (C2). Fix the auxiliary registration imaging lens (C2) at the center of the front of the ultraviolet imaging lens (UV1) using a positioning support to achieve visual centroidal registration between the auxiliary registration imaging lens (C2) and the ultraviolet imaging lens (UV1).

[0015] b) Under the premise of ensuring that the images captured by the main visible light imaging lens (C1) and the auxiliary registration imaging lens (C2) overlap, image data of the main visible light imaging lens (C1) and the auxiliary registration imaging lens (C2) are acquired respectively.

[0016] c) Using the already trained DeepLabv3Plus model, extract the high-level feature maps of the image data acquired by the main visible light imaging lens (C1) and the image data acquired by the auxiliary registration imaging lens (C2), denoted as F1 and F2 respectively;

[0017] d) Perform Fourier transforms on F1 and F2 respectively to obtain the frequency domain feature maps of the higher-level feature maps, denoted as A1 and A2 respectively. The frequency domain feature maps contain the amplitude and phase information of the higher-level feature layers.

[0018] e) Calculate the cross power spectrum of A1 and A2, and then perform inverse Fourier transform. By searching for peaks, obtain the relative translation of the two images, that is, obtain the relative translation of the main visible light imaging lens (C1) and the auxiliary registration imaging lens (C2).

[0019] f) The frequency domain feature map can be divided into an amplitude spectrum map and a phase spectrum map. Logarithmic polar coordinate transformation is performed on the amplitude spectrum map of A1 and the amplitude spectrum map of A2 respectively to obtain the logarithmic polar coordinate image of the amplitude spectrum map, which is denoted as L1 and L2 respectively.

[0020] g) Calculate the cross-correlation function of L1 and L2, obtain the point of maximum value, extract the polar angle information of the maximum value point, that is, obtain the relative rotation angle between the main visible light imaging lens (C1) and the auxiliary registration imaging lens (C2).

[0021] h) Equivalent the relative translation and relative rotation angle information of the main visible light imaging lens (C1) and the auxiliary registration imaging lens (C2) to the relative translation and relative rotation angle information of the main visible light imaging lens (C1) and the ultraviolet imaging lens (UV1), and calculate the registration information by combining the field of view information of the main visible light imaging lens (C1) and the ultraviolet imaging lens (UV1).

[0022] Further, in step a: the positioning support includes a rigid fixing part and an extension support part. The extension support part extends horizontally outward from one side of the rigid fixing part. The rigid fixing part is detachably connected to the main structure of the ultraviolet imaging lens (UV1). The auxiliary registration imaging lens (C2) is fixed at the end of the extension support part, so that the auxiliary registration imaging lens (C2) is located at the center of the front of the ultraviolet imaging lens (UV1), and the auxiliary registration imaging lens (C2) is as close as possible to the ultraviolet imaging lens (UV1).

[0023] Furthermore, the method for placing the auxiliary registration imaging lens (C2) at the center position directly in front of the ultraviolet imaging lens (UV1) in step a is as follows: the field of view of the auxiliary registration imaging lens (C2) and the ultraviolet imaging lens (UV1) is calibrated by using a multi-point ultraviolet light source calibration method.

[0024] Furthermore, in step b: ensure that the main visible light imaging lens (C1) and the auxiliary registration imaging lens (C2) have an overlap of not less than 50%.

[0025] Furthermore, in step c: the DeepLabv3Plus model is a pre-trained deep neural network model. When deploying the deep neural network model, it is necessary to remove the last few layers of the network so that the network can output image features.

[0026] Another objective of this invention is to provide a field-of-view registration system for ultraviolet lens imaging and visible light lens imaging, comprising a main visible light imaging lens (C1), an auxiliary registration imaging lens (C2), and an ultraviolet imaging lens (UV1). The auxiliary registration imaging lens (C2) is smaller than the ultraviolet imaging lens (UV1). The auxiliary registration imaging lens (C2) is fixed to the center of the front of the ultraviolet imaging lens (UV1) by a positioning support (10). The auxiliary registration imaging lens (C2) and the ultraviolet imaging lens (UV1) are integrated. The main visible light imaging lens (C1), the auxiliary registration imaging lens (C2), and the ultraviolet imaging lens (UV1) are arranged in parallel.

[0027] The positioning support (10) includes a rigid fixing part (101) and an extension support part (102). The extension support part (102) extends horizontally outward from one side of the rigid fixing part (101). The rigid fixing part (101) is connected to the main structure of the ultraviolet imaging lens (UV1), and the extension support part (102) is connected to the auxiliary registration imaging lens (C2).

[0028] Another object of the present invention is to provide a combined lens, including an ultraviolet imaging lens (UV1) and an auxiliary registration imaging lens (C2), wherein the size of the auxiliary registration imaging lens (C2) is smaller than that of the ultraviolet imaging lens (UV1), and the auxiliary registration imaging lens (C2) is disposed at the center of the front of the ultraviolet imaging lens (UV1), wherein the combined lens is configured to perform the field registration method described above.

[0029] Another object of the present invention is to provide a positioning support member, including a rigid fixing part (101) and an extension support part (102), the extension support part (102) extending horizontally outward from the middle side of the rigid fixing part (101), wherein the positioning support member is configured to assemble a combined lens as described above, the rigid fixing part (101) is used to connect an ultraviolet imaging lens (UV1), and the extension support part (102) is used to connect an auxiliary registration imaging lens (C2).

[0030] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention adds a miniature lens as an auxiliary registration imaging lens, which indirectly realizes the registration of the ultraviolet imaging field of view and the visible light imaging field of view, so that users can not only detect ultraviolet signals, but also accurately locate the signal source. Attached Figure Description

[0031] Figure 1 A schematic diagram showing the spectral comparison of ultraviolet and visible light from corona discharge is provided.

[0032] Figure 2 This diagram illustrates the detection principle of current ultraviolet imagers.

[0033] Figure 3 This is a schematic diagram of the rigid fixation of the auxiliary registration imaging lens (C2) and the ultraviolet imaging lens (UV1) of the present invention;

[0034] Figure 4 This is a simplified schematic diagram of a field-of-view registration system for ultraviolet lens imaging and visible light lens imaging according to the present invention.

[0035] Figure 5 This is a simplified schematic diagram of the multi-point ultraviolet light source calibration method of the present invention;

[0036] Figure 6 This is a neural network structure diagram of the DeepLabv3Plus model described in this invention;

[0037] The relevant markings in the attached diagram are: C1 - main visible light imaging lens, C2 - auxiliary registration imaging lens, UV1 - ultraviolet imaging lens, 10 - positioning support, 101 - rigid fixing part, 102 - extension support part. Detailed Implementation

[0038] The technical solutions of the embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0039] The field-of-view registration method for ultraviolet lens imaging and visible light lens imaging provided in this application embodiment has the following specific implementation steps:

[0040] The first preparatory work is to build a registration system:

[0041] Select a primary visible light imaging lens (C1), select an ultraviolet imaging lens (UV1), and select a miniature lens as an auxiliary registration imaging lens (C2). Fix the auxiliary registration imaging lens (C2) in the center of the front of the ultraviolet imaging lens (UV1) using a positioning support to achieve visual centroidal registration between the auxiliary registration imaging lens (C2) and the ultraviolet imaging lens (UV1). In other words, ensure that the centers of the auxiliary registration imaging lens (C2) and the ultraviolet imaging lens (UV1) are on the same straight line.

[0042] The selection principle for this miniature lens is firstly to ensure its small size so that it will not cause unnecessary imaging interference when placed close to other imaging lenses. Another principle is to ensure its high image quality, requiring it to have sufficient image resolution and spatial resolution. A miniature image acquisition lens with a resolution of 1280*720 can be selected. When selecting, the ISO12233 resolution test chart is used to test to ensure that the resolution requirements are met.

[0043] like Figure 3 As shown, the positioning support 10 described in this embodiment presents an overall "T" shaped structure. It includes a rigid fixing part 101 and an extension support part 102. The extension support part extends horizontally outward from the middle side of the rigid fixing part. The rigid fixing part 101 can be connected to the main structure of the ultraviolet imaging lens (UV1) in a detachable connection form (such as screw connection). The positioning support part can be made of aluminum alloy, magnesium alloy, etc., in one piece, or it can be made of PCB and covered with a thicker copper sheet on the other side.

[0044] The auxiliary registration imaging lens (C2) is fixed to the end of the extension support 102, so that the auxiliary registration imaging lens (C2) is located in the center of the front of the ultraviolet imaging lens (UV1), and the auxiliary registration imaging lens (C2) is as close as possible to the glass of the ultraviolet imaging lens (UV1).

[0045] After ensuring that the visual centers of the auxiliary registration imaging lens (C2) and the ultraviolet imaging lens (UV1) are aligned, their relative positions cannot be easily changed. Therefore, it is necessary to perform visual center calibration of the auxiliary registration imaging lens (C2) and the ultraviolet imaging lens (UV1). In this embodiment, the field of view of the auxiliary registration imaging lens (C2) and the ultraviolet imaging lens (UV1) will be calibrated by using the multi-point ultraviolet light source calibration method.

[0046] The multi-point ultraviolet light source calibration method involves setting up multiple ultraviolet light sources and distributing them according to certain rules at a certain distance directly in front of the ultraviolet imaging lens (UV1). The ultraviolet imaging spot is then aligned with the visible light source position of the auxiliary registration imaging lens (C2) to achieve visual center coincidence. Specifically, as follows... Figure 5As shown, a planar three-dimensional support structure is generally designed, and multiple ultraviolet light sources (alcohol lamps can be used as ultraviolet light sources) are arranged on the support according to a certain pattern. The planar three-dimensional support is installed at a distance of about 5 to 10 meters from the ultraviolet imaging lens (UV1) to ensure that all the flames of the alcohol lamps can be seen in the image of the auxiliary registration imaging lens (C2), and a stable light spot can be seen in the image of the ultraviolet imaging lens (UV1). In this way, the visual center can be coincident.

[0047] After completing the above preparations, the next step is to register the main visible light imaging lens (C1) with the ultraviolet imaging lens (UV1).

[0048] like Figure 4 As shown, under certain constraints, the images captured by the main visible light imaging lens (C1) and the auxiliary registration imaging lens (C2) are guaranteed to have an overlap of no less than 50%, that is, to ensure that they are captured in a nearly parallel manner from top to bottom. Image data from the main visible light imaging lens (C1) and the auxiliary registration imaging lens (C2) are acquired separately, and high-level feature maps of the images from the main visible light imaging lens (C1) and the auxiliary registration imaging lens (C2) are extracted using a pre-trained DeepLabv3Plus, denoted as F1 and F2 respectively. F1 and F2 can also be considered as two incomprehensible images. Figure 6 The diagram shows the DeepLabv3Plus neural network structure. DeepLabv3Plus is a pre-trained deep neural network model. Its training data includes images acquired by C1 and C2. The training data uses an alternating comparison method between C1 and C2. The training loss incorporates the relative scaling, offset, and rotation factors of the C1 and C2 images. Furthermore, when deploying the deep neural network model, it is necessary to remove some of the last layers of the network so that the network can output image features.

[0049] Perform Fourier transforms on the high-level feature maps F1 and F2 respectively to obtain the frequency domain feature maps of the high-level feature maps, denoted as A1 and A2 respectively. The frequency domain feature maps contain the amplitude and phase information of the high-level feature layers, and the calculation method can be referred to the following formula:

[0050]

[0051] Cross-power spectrum calculation is performed on frequency domain feature maps A1 and A2, followed by inverse Fourier transform (phase correlation). The relative translation ω between the two images is obtained by searching for peaks, which gives the relative translation between the main visible light imaging lens (C1) and the auxiliary registration imaging lens (C2). The cross-power spectrum calculation can be referenced from the following formula:

[0052] C fg (u,v)=F(u,v)·G * (u,v)

[0053] Where F and G are the Fourier transforms of two frequency domain feature maps, and G * (u,v) is the complex conjugate of the Fourier transform.

[0054] The inverse Fourier transform can be calculated using the following formula:

[0055]

[0056] The frequency domain feature map can be divided into an amplitude spectrum and a phase spectrum. Logarithmic polar coordinate transformation is performed on the amplitude spectrum of A1 and the amplitude spectrum of A2 respectively to obtain the logarithmic polar coordinate images of the amplitude spectrum, denoted as L1 and L2 respectively. The horizontal axis of the logarithmic polar coordinate image represents the polar-logarithmic distance, and the vertical axis represents the polar angle.

[0057] Cross-correlation calculations are performed on images L1 and L2 to obtain their vertical translation h. The relative rotation of A1 and A2 is obtained through transformation. In detail, the cross-correlation function of L1 and L2 is calculated to obtain the point of maximum value. The polar angle information of the maximum value point is extracted, that is, the relative rotation angle between the main visible light imaging lens (C1) and the auxiliary registration imaging lens (C2) is obtained.

[0058] The relative translation and relative rotation angle information of the main visible light imaging lens (C1) and the auxiliary registration imaging lens (C2) are equivalent to the relative translation and relative rotation angle information of the main visible light imaging lens (C1) and the ultraviolet imaging lens (UV1). The registration information is calculated by combining the field of view information of the main visible light imaging lens (C1) and the ultraviolet imaging lens (UV1).

[0059] Based on the above preparations, it can be seen that the visual centers of the auxiliary registration imaging lens (C2) and the ultraviolet imaging lens (UV1) are considered to coincide. After the above registration work, the visual centers of the main visible light imaging lens (C1) and the auxiliary registration imaging lens (C2) are also registered to coincide. Therefore, it can be seen that the main visible light imaging lens (C1) and the ultraviolet imaging lens (UV1) are indirectly registered to coincide.

[0060] Given the field of view of the ultraviolet imaging lens (UV1) (determined by the photosensitive element) and the field of view of the main visible light imaging lens (C1), pixel-level field of view registration information can be further achieved by performing a certain scaling on the ultraviolet imaging lens (UV1).

[0061] Thus, by acquiring images from the main visible light imaging lens (C1) and the ultraviolet imaging lens (UV1) in real time, and combining them with the relative translation information and relative rotation angle information obtained in the above steps, and based on image fusion technology, real-time simultaneous display of visible light imaging and ultraviolet imaging after registration can be achieved.

[0062] The above descriptions are merely some embodiments of the present invention. Those skilled in the art can make various modifications and improvements without departing from the inventive concept of the present invention, and these all fall within the scope of protection of the present invention.

Claims

1. A method for field-of-view registration between ultraviolet lens imaging and visible light lens imaging, characterized in that, Includes the following steps: a Select a main visible light imaging lens (C1), select an ultraviolet imaging lens (UV1), and select a miniature lens as an auxiliary registration imaging lens (C2). Fix the auxiliary registration imaging lens (C2) at the center of the front of the ultraviolet imaging lens (UV1) through the positioning support (10) to achieve visual centroidal registration between the auxiliary registration imaging lens (C2) and the ultraviolet imaging lens (UV1). b Under the premise of ensuring that the images captured by the main visible light imaging lens (C1) and the auxiliary registration imaging lens (C2) overlap, image data of the main visible light imaging lens (C1) and the auxiliary registration imaging lens (C2) are collected respectively. c Using the pre-trained DeepLabv3Plus model, high-level feature maps were extracted from the image data acquired by the main visible light imaging lens (C1) and the image data acquired by the auxiliary registration imaging lens (C2), which were denoted as F1 and F2, respectively. d Perform Fourier transforms on F1 and F2 respectively to obtain the frequency domain feature maps of the higher-level feature maps, denoted as A1 and A2 respectively. The frequency domain feature maps contain the amplitude and phase information of the higher-level feature layers. e The cross-power spectrum of A1 and A2 is calculated, and then an inverse Fourier transform is performed. The relative translation of the two images is obtained by searching for peaks. ， That is, to obtain the relative translation between the main visible light imaging lens (C1) and the auxiliary registration imaging lens (C2); f The frequency domain feature map can be divided into an amplitude spectrum and a phase spectrum. Logarithmic polar coordinate transformation is performed on the amplitude spectrum of A1 and the amplitude spectrum of A2 respectively to obtain the logarithmic polar coordinate images of the amplitude spectrum, which are denoted as L1 and L2 respectively. g Calculate the cross-correlation function of L1 and L2, obtain the point of maximum value, extract the polar angle information of the maximum value point, that is, obtain the relative rotation angle between the main visible light imaging lens (C1) and the auxiliary registration imaging lens (C2); h The relative translation and relative rotation angle information of the main visible light imaging lens (C1) and the auxiliary registration imaging lens (C2) are equivalent to the relative translation and relative rotation angle information of the main visible light imaging lens (C1) and the ultraviolet imaging lens (UV1). The registration information is calculated by combining the field of view information of the main visible light imaging lens (C1) and the ultraviolet imaging lens (UV1). i By acquiring images in real time from the main visible light imaging lens (C1) and the ultraviolet imaging lens (UV1), and combining them with the steps... h The relative translation and relative rotation angle information obtained can be used to achieve real-time simultaneous display of visible light imaging and ultraviolet imaging after registration, based on image fusion technology.

2. The field-of-view registration method for ultraviolet lens imaging and visible light lens imaging according to claim 1, characterized in that, The steps a In the middle: the positioning support (10) includes a rigid fixing part (101) and an extension support part (102). The extension support part (102) extends horizontally outward from one side of the rigid fixing part (101). The rigid fixing part (101) is detachably connected to the main structure of the ultraviolet imaging lens (UV1). The auxiliary registration imaging lens (C2) is fixed at the end of the extension support part (102), so that the auxiliary registration imaging lens (C2) is located at the center of the front of the ultraviolet imaging lens (UV1), and the auxiliary registration imaging lens (C2) is as close as possible to the ultraviolet imaging lens (UV1).

3. The field-of-view registration method for ultraviolet lens imaging and visible light lens imaging according to claim 2, characterized in that, The steps a The method to achieve the center position of the auxiliary registration imaging lens (C2) directly in front of the ultraviolet imaging lens (UV1) is as follows: the field of view of the auxiliary registration imaging lens (C2) and the ultraviolet imaging lens (UV1) is calibrated by using a multi-point ultraviolet light source calibration method.

4. The field-of-view registration method for ultraviolet lens imaging and visible light lens imaging according to claim 1, characterized in that, The steps b In the middle: Ensure that the images captured by the main visible light imaging lens (C1) and the auxiliary registration imaging lens (C2) have an overlap of no less than 50%.

5. The field-of-view registration method for ultraviolet lens imaging and visible light lens imaging according to claim 1, characterized in that, The steps c In Chinese: The DeepLabv3Plus model is a pre-trained deep neural network model. When deploying a deep neural network model, it is necessary to remove the last few layers of the network so that the network can output image features.

6. A combination lens, characterized in that, The lens includes an ultraviolet imaging lens (UV1) and an auxiliary registration imaging lens (C2), wherein the size of the auxiliary registration imaging lens (C2) is smaller than that of the ultraviolet imaging lens (UV1), and the auxiliary registration imaging lens (C2) is positioned at the center of the front of the ultraviolet imaging lens (UV1). The combined lens is configured to perform the field registration method as described in any one of claims 1-5.

7. A positioning support member, characterized in that, It includes a rigid fixing part (101) and an extension support part (102), the extension support part (102) extending horizontally outward from the middle side of the rigid fixing part (101), wherein the positioning support is configured to assemble the combined lens as described in claim 6, the rigid fixing part (101) is used to connect the ultraviolet imaging lens (UV1), and the extension support part (102) is used to connect the auxiliary registration imaging lens (C2).

Images