Object detection method and photographic apparatus therefor

Abstract

The present application discloses an object detection method and a photographic device, which can effectively improve the recognition accuracy of the photographic device. The method comprises the following steps: obtaining a first reference image and a second reference image covering a specific monitoring area by using a first exposure parameter and a second exposure parameter respectively; obtaining a first detection image and a second detection image covering the specific monitoring area during the object detection by using the first exposure parameter and the second exposure parameter respectively; calculating a first pixel variation between the first reference image and the first detection image, and a second pixel variation between the second reference image and the second detection image; and analyzing the first pixel variation and the second pixel variation to determine whether a target object is located in the specific monitoring area. The present application adopts a parallel object detection mode of short exposure images and long exposure images, and selectively adopts the object detection mode of short exposure images or long exposure images under the condition of meeting specific conditions, so as to balance the recognition accuracy and power saving performance.

Description

Technical Field

[0001] This invention relates to an object detection method and photographic equipment, and particularly to an object detection method and photographic equipment that can effectively improve the accuracy of identification. Background Technology

[0002] Traditional surveillance cameras acquire reference and detection images using a single exposure parameter. The reference image is captured when no target object is present within the camera's monitoring range; the detection image is captured during the camera's operation to determine if a target object has entered the monitoring range. Traditional surveillance cameras calculate the pixel difference between corresponding positions in the reference and detection images. If the pixel difference in a certain area is below or equal to a specific threshold, it indicates no significant pixel change in that area, meaning no target object is present. If the pixel difference in a certain area is above the threshold, it indicates significant pixel change in that area, and therefore a target object is considered present. However, if the ambient brightness within the monitoring range is low, and the target object itself does not have its own light source, the surveillance camera will struggle to identify the target object blending into a dark background in the detection image. Therefore, designing a surveillance camera capable of accurately identifying target objects in low-light environments has become a key development goal for the surveillance industry. Summary of the Invention

[0003] This invention relates to an object detection method and photographic equipment that can effectively improve the accuracy of object identification.

[0004] The present invention further discloses an object detection method that can effectively improve the accuracy of identification, comprising: acquiring a first reference image and a second reference image covering a specific monitoring area during a non-detection period using a first exposure parameter and a second exposure parameter greater than the first exposure parameter; acquiring a first detection image and a second detection image covering the specific monitoring area during an object detection period using the first exposure parameter and the second exposure parameter; calculating a first pixel variation between the first reference image and the first detection image, and a second pixel variation between the second reference image and the second detection image; and analyzing the first pixel variation and the second pixel variation to determine whether the target object is located within the specific monitoring area.

[0005] The present invention also discloses that the object detection method further includes setting a region of interest within the first reference image and the first detection image, so as to calculate the first pixel change within the region of interest.

[0006] The present invention also discloses that the object detection method further includes stopping the capture of related reference images and detection images using the second exposure parameter if the target object determined by the first pixel change amount matches the target object determined by the second pixel change amount.

[0007] The present invention also discloses that the object detection method further includes dividing the first range where the target object is located and the second range outside the target object in the first pixel variation, comparing the difference between the first range and the second range with a gradient threshold, and stopping the capture of related reference images and detection images using the second exposure parameters if the difference exceeds the gradient threshold.

[0008] The present invention also discloses that the object detection method further includes activating a light detector to obtain ambient light information, and analyzing the ambient light information to determine whether to extract the second exposure parameters to capture the second reference image and the second detection image.

[0009] The present invention also discloses that the object detection method further includes analyzing the brightness information of the ambient light information to calculate the second exposure parameter accordingly.

[0010] The present invention also discloses that the object detection method further includes comparing the brightness calculation value of the first reference image with a brightness threshold, and if the brightness calculation value is lower than the brightness threshold, extracting the second exposure parameter to capture the second reference image and the second detection image.

[0011] The present invention also discloses that the object detection method further includes obtaining one of the average brightness value and the maximum brightness value of the first reference image, and transforming the calculated value using a preset weight to calculate the second exposure parameter accordingly.

[0012] The present invention also discloses that the object detection method further includes obtaining the brightness histogram distribution information of the first reference image, dividing the brightness histogram distribution information into a high brightness distribution area and a low brightness distribution area, and calculating the second exposure parameter accordingly using the distribution ratio of the high brightness distribution area and the low brightness distribution area.

[0013] The present invention also discloses that the object detection method further includes analyzing the change in the second pixel to determine whether to adjust the second exposure parameter.

[0014] The present invention also discloses a photographic device with high recognition accuracy, including an image acquisition unit and a processing unit. The image acquisition unit acquires a first reference image and a second reference image covering a specific monitoring area during non-detection periods using a first exposure parameter and a second exposure parameter greater than the first exposure parameter; and acquires a first detection image and a second detection image covering the specific monitoring area during object detection periods using the first exposure parameter and the second exposure parameter. The processing unit is electrically connected to the image acquisition unit. The processing unit calculates a first pixel variation between the first reference image and the first detection image, and a second pixel variation between the second reference image and the second detection image, and analyzes the first pixel variation and the second pixel variation to determine whether a target object is located within the specific monitoring area.

[0015] The photographic device and object detection method of the present invention preferably utilize both short-exposure images (i.e., a first reference image and a first detection image generated with a first exposure parameter) and long-exposure images (i.e., a second reference image and a second detection image generated with a second exposure parameter) simultaneously to determine the presence of a target object. Relying solely on short-exposure images may fail to correctly detect low-contrast target objects in low-light environments; relying solely on long-exposure images may result in oversaturation in certain parts of the image due to the bright characteristics of the target object, making detection and judgment difficult. Therefore, the present invention employs a parallel object detection mode using both short-exposure and long-exposure images, and simultaneously provides an automatic off and on function for multiple exposures. Under specific conditions, either short-exposure or long-exposure image object detection modes can be selectively used to balance recognition accuracy and power saving efficiency. Attached Figure Description

[0016] Figure 1 This is a functional block diagram of a photographic device according to an embodiment of the present invention.

[0017] Figure 2 This is a schematic diagram of the monitoring area provided by the photographic device in an embodiment of the present invention.

[0018] Figure 3 This is a flowchart of an object detection method according to an embodiment of the present invention.

[0019] Figures 4 to 7 This is a schematic diagram of images obtained by the photographic device of this invention under different conditions according to an embodiment of the invention.

[0020] Figure 8 and Figure 9 This is a schematic diagram of the image obtained by the photographic device according to an embodiment of the present invention after image analysis.

[0021] Figure 10 This is a schematic diagram of the brightness histogram distribution information of the first reference image in an embodiment of the present invention.

[0022] The reference numerals in the attached figures are explained as follows:

[0023] 10. Photography equipment

[0024] 12 Image Acquisition Device

[0025] 14. Processing Unit

[0026] 16 Light Detectors

[0027] Ot target object

[0028] R Region of Interest

[0029] I1r First Reference Image

[0030] I2r Second Reference Image

[0031] I1d First Detected Image

[0032] I2d second detection image

[0033] NS_REF First reference pixel array

[0034] NS_CMP First Detection Pixel Array

[0035] D_NS First pixel array difference

[0036] LS_REF Second Reference Pixel Array

[0037] LS_CMP Second Detection Pixel Array

[0038] D_LS second pixel array difference

[0039] M1 First Marker Box

[0040] M2 Second Marker Box

[0041] A1 High brightness distribution area

[0042] A2 Low brightness distribution area

[0043] Steps S100, S102, S104, S106, S108, S110 Detailed Implementation

[0044] Please see Figure 1 and Figure 2 , Figure 1 This is a functional block diagram of the photographic device 10 according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the monitoring area provided by the photographic device 10 according to an embodiment of the present invention. The photographic device 10 may include an image acquisition unit 12 and a processing unit 14 electrically connected to each other. The image acquisition unit 12 can be used to acquire reference images and detection images related to the monitoring area of ​​the photographic device 10; the image acquisition unit 12 may have its own image capturing function, or it may be connected to another image capture unit to acquire the images generated therefrom. The processing unit 14 may be connected to the image acquisition unit 12 via wired or wireless means, thereby analyzing and judging the reference images and detection images. The processing unit 14 may be a computing unit independent of the image acquisition unit 12, or it may be a computing module built into the image acquisition unit 12.

[0045] Generally, the camera device 10 is not limited to being installed in indoor or outdoor spaces. If the monitored area has significant changes in brightness, the camera device 10 can use different exposure parameters to obtain reference and detection images within the same field of view. Then, based on the pixel variation of the reference and detection images under each exposure parameter, it can determine whether a target object Ot appears in the monitored area. Different exposure parameters typically include normal and slightly higher exposure parameters. Slightly higher exposure parameters can be used to obtain overexposed reference and detection images, thereby improving the accuracy of identifying low-contrast target objects Ot within the monitored area. Low-contrast target objects Ot mainly refer to dark objects against a dark background, but practical applications are not limited to this. Different exposure parameters can originate from the system settings of the camera device 10, be set according to changes in the ambient brightness of the monitored area, or be adjusted based on the image characteristics of a specific image.

[0046] like Figure 2 As shown, the camera device 10 may be installed inside an office building. Due to usage requirements, the office building may have a long corridor extending into the distance. Only some of the office building's lighting is on. When a target object Ot (e.g., a pedestrian) gradually approaches the camera device 10 from the dark corridor, the dimly lit office serves as the foreground's dark background, and the pedestrian approaching the camera device 10 from the darkness is considered a dark object. If the camera device 10 only uses a reference image and a detection image obtained with a single exposure parameter for analysis, it will be difficult to clearly identify the target object Ot in the image obtained with normal exposure parameters. Therefore, the camera device 10 can automatically or manually activate the multiple exposure function when trigger conditions are met, in order to correctly identify whether the target object Ot exists in the monitored area.

[0047] Therefore, the photographic device 10 may optionally include a light detector 16 electrically connected to the processing unit 14. The photographic device 10 can use the light detector 16 to obtain ambient light information of the monitored area, and thereby determine how to set the difference between normal exposure parameters and overexposure parameters. If the photographic device 10 does not have a light detector, or although it has a light detector 16, it has not been activated due to the lack of triggering conditions, then the difference between normal exposure parameters and overexposure parameters can be determined based on the pixel brightness of the first image obtained (usually a reference image, but practical applications are not limited to this). In this way, the photographic device 10 of the present invention can automatically or manually activate the multiple exposure function, and by analyzing the pixel variation between the reference image and the detected image obtained with different exposure parameters, it can correctly and effectively improve the recognition accuracy of the target object Ot under various ambient brightness conditions.

[0048] It is worth mentioning that the multiple exposure function of the present invention refers to obtaining corresponding reference images and detection images using two or more exposure parameters respectively. The following description will use two different exposure parameters as preferred embodiments; however, the number of variations in exposure parameters is not limited to this and depends on the settings or actual needs. Therefore, other possible variations for capturing images using three or more exposure parameters for target object Ot detection will not be described in detail. Furthermore, the setting and adjustment of exposure parameters can be performed from the camera exposure time, shutter speed, aperture size, or circuit gain of the photographic device 10; the actual selection depends on design requirements.

[0049] Please see Figures 2 to 9 , Figure 3 This is a flowchart of the object detection method according to an embodiment of the present invention. Figures 4 to 7 This is a schematic diagram of images obtained by the photographic device 10 under different conditions according to an embodiment of the present invention. Figure 8 and Figure 9 This is a schematic diagram of the image obtained by the photographic device 10 according to an embodiment of the present invention after image analysis. Figure 3 The object detection method described above can be applied to: Figure 1 and Figure 2 The photographic device 10 is shown. Regarding the object detection method, firstly, step S100 is executed to obtain a first exposure parameter and a second exposure parameter greater than the first exposure parameter. The first exposure parameter and the second exposure parameter may be preset values ​​of the photographic device 10, or variable values ​​calculated based on ambient light information from the light detector 16 or image features of a specific image. Next, steps S102 and S104 are executed to obtain a first reference image I1r and a second reference image I2r covering a specific monitoring area during non-detection periods using the first exposure parameter and the second exposure parameter, and to obtain a first detection image I1d and a second detection image I2d covering a specific monitoring area during object detection using the first exposure parameter and the second exposure parameter.

[0050] Next, step S106 can be selectively executed to define a region of interest (ROI) R within the acquired image that corresponds to the area where the target object Ot has a high probability of appearing. For example, ROI R could correspond to a corridor in an office building; therefore, the first reference image I1r, the second reference image I2r, the first detection image I1d, and the second detection image I2d would all have ROI R in the same location and within the same range. Next, step S108 is executed to calculate the first pixel variation between the first reference image I1r and the first detection image I1d, and the second pixel variation between the second reference image I2r and the second detection image I2d. In step S108, the object detection method of the present invention may calculate the pixel variation between all pixels in the reference image and the detection image, or it may calculate the pixel variation between the ROI R in the reference image and the ROI R in the detection image; the actual application depends on the design requirements. Finally, step S110 is executed to analyze and compare the first pixel variation and the second pixel variation to determine whether the target object Ot is present within a specific monitoring area.

[0051] like Figure 8 and Figure 9 As shown, the region of interest (ROI) R of the first reference image I1r provides a first reference pixel array NS_REF, and the region of interest R of the first detected image I1d provides a first detected pixel array NS_CMP; the region of interest R of the second reference image I2r also provides a second reference pixel array LS_REF, and the region of interest R of the second detected image I2d provides a second detected pixel array LS_CMP. In this embodiment, step S108 calculates the first pixel variation between the first reference image I1r and the first detected image I1d, which can be obtained from the difference between the first reference pixel array NS_REF and the first detected pixel array NS_CMP to obtain the first pixel array difference D_NS. Then, the second pixel variation between the second reference image I2r and the second detected image I2d is calculated, which can be obtained from the difference between the second reference pixel array LS_REF and the second detected pixel array LS_CMP to obtain the second pixel array difference D_LS. Figure 8 and Figure 9 It can be observed that some regions within the first pixel array D_NS difference have large pixel variations (meaning they exceed a certain threshold), and the pixels in these regions are assigned to the first bounding box M1; similarly, some regions within the second pixel array difference D_LS also have large pixel variations (meaning they exceed a certain threshold), and the pixels in these regions are assigned to the second bounding box M2.

[0052] The first marker box M1 serves as the basis for the photographic device 10 to analyze the first reference image I1r and the first detection image I1d to determine whether the target object Ot is located within the monitoring area. The second marker box M2 serves as the basis for the photographic device 10 to analyze the second reference image I2r and the second detection image I2d to determine whether the target object Ot is located within the monitoring area. If the relative positions and sizes of the first marker box M1 and the second marker box M2 in the first detection image I1d and the second detection image I2d are the same or similar, it can be considered that the photographic device 10 has captured the same target object Ot in the reference image and the detection image obtained using different exposure parameters.

[0053] In step S110, the first pixel change and the second pixel change may be compared to the same change threshold value, or the first pixel change and the second pixel change may be compared to different change threshold values. If the first pixel change (or the first marker box M1) and the second pixel change (or the second marker box M2) are both greater than or equal to the change threshold value, it indicates that the target object Ot appears in the monitoring area of ​​the imaging device 10 and is simultaneously captured by the first detection image I1d generated by the first exposure parameter and the second detection image I2d generated by the second exposure parameter. If the first pixel change (or the first marker box M1) and the second pixel change (or the second marker box M2) are both less than the change threshold, it means that there is no target object in the monitoring area of ​​the camera device 10. It is also possible that neither the first detection image I1d generated by the first exposure parameter nor the second detection image I2d generated by the second exposure parameter can capture the target object. In this case, a third reference image and a third detection image can be selectively generated with a third exposure parameter that is higher than the first and second exposure parameters, thereby confirming that the target object does not exist in the monitoring area of ​​the camera device 10, or that it cannot be captured by the second detection image I2d due to low ambient brightness.

[0054] If the first pixel change (or the first marker box M1) is less than the change threshold value, but the second pixel change (or the second marker box M2) is greater than or equal to the change threshold value, it means that the target object Ot cannot be read in the first reference image I1r and the first detection image I1d obtained using the first exposure parameters because the ambient brightness is too dark. However, the target object Ot can still be detected in the second reference image I2r and the second detection image I2d obtained using the second exposure parameters. Therefore, the object detection method can determine that the target object Ot appears in the monitoring area of ​​the camera device 10. If the change in the first pixel (or the first marker box M1) is greater than or equal to the change threshold value, and the change in the second pixel (or the second marker box M2) is less than the change threshold value, it may be because the ambient brightness is too high or other factors cause the second reference image I2r and the second detection image I2d to have difficulty correctly detecting the target object Ot. In this case, the object detection method of the present invention can still use the first reference image I1r and the first detection image I1d to determine that the target object Ot appears in the monitoring area of ​​the camera device 10. The value of the change threshold value depends on the optical parameters of the camera device 10, and will not be described in detail here.

[0055] If both the first pixel change and the second pixel change can be used to determine the target object Ot (e.g., both exceed the change threshold value), and the target object Ot determined by the first pixel change matches the target object Ot determined by the second pixel change, it means that the camera device 10 can correctly determine whether the target object Ot appears in the monitoring area by using the first reference image I1r and the first detection image I1d obtained by the first exposure parameters. In this state, the object detection method of the present invention can selectively turn off the multiple exposure function, that is, stop using the second exposure parameters to obtain the second reference image I2r and the second detection image I2d, so as to save energy consumption, system computing power and information storage.

[0056] Alternatively, the object detection method may further divide the first pixel array difference D_NS of the first pixel variation into a first range (i.e., within the first marker box M1) and a second range (i.e., outside the first marker box M1) where the target object Ot is located, and then compare the difference between the first range and the second range with a preset gradient threshold. The value of the gradient threshold depends on the ambient brightness of the imaging device 10, and therefore will not be described in detail here. If the difference is lower than or equal to the gradient threshold, it means that the first marker box M1 is difficult to confirm the existence of the target object Ot. In this case, object detection must still be performed using the reference image and the detection image obtained with the first exposure parameters and the second exposure parameters. If the difference is greater than the gradient threshold, it means that the first marker box M1 is sufficient to confirm the existence of the target object Ot. The target object Ot can be clearly identified using the first reference image I1r and the first detection image I1d. In this state, the object detection method of the present invention can selectively turn off the multiple exposure function, that is, stop using the second exposure parameters to obtain the second reference image I2r and the second detection image I2d.

[0057] After the imaging device 10 and its object detection method disable the multiple exposure function, the light detector 16 can be periodically activated to obtain ambient light information of the monitored area at any time, and the ambient light information can be analyzed in real time to determine whether to restart the multiple exposure function. For example, if the ambient light information is higher than the preset brightness threshold, it means that the ambient lighting of the imaging device 10 is sufficient, and the multiple exposure function does not need to be turned on. If the ambient light information is lower than or equal to the preset brightness threshold, it means that the environment where the imaging device 10 is located is dark, and the multiple exposure function needs to be turned on to ensure that the imaging device 10 can correctly detect whether the target object Ot appears in the monitored area. The second exposure parameter of the multiple exposure function can be calculated based on the brightness information of the ambient light information, or it can be determined based on the system preset value of the imaging device 10. The value of the brightness threshold can be determined according to the ambient brightness conditions of the environment where the imaging device 10 is located, so it will not be explained in detail here.

[0058] Alternatively, after the imaging device 10 and its object detection method disable the multiple exposure function, they can selectively analyze the image brightness of the first reference image I1r and / or the first detection image I1d to determine whether to restart the multiple exposure function. The image brightness can be the average or maximum pixel brightness of the entire image, or it can be a brightness calculation value obtained by using the average and maximum pixel brightness. For example, if the image brightness of the first reference image I1r or the first detection image I1d is higher than a preset brightness threshold, it indicates that the ambient lighting of the imaging device 10 is sufficient, and there is no need to enable the multiple exposure function. If the brightness of the first reference image I1r or the first detected image I1d is lower than or equal to a preset brightness threshold, it indicates that the environment in which the camera device 10 is located is too dark, and in this case, it is necessary to enable the multiple exposure function. The second exposure parameter of the multiple exposure function can be determined according to the brightness of the first reference image I1r or the first detected image I1d. For example, the average pixel brightness and the maximum pixel brightness of the first reference image I1r or the first detected image I1d can be calculated first, and then the second exposure parameter can be calculated based on the average pixel brightness, the maximum pixel brightness, or the brightness calculation value obtained by the average pixel brightness and the maximum pixel brightness, with a preset weight. The calculation method of the brightness calculation value and the weight of the preset weight can be set according to the system of the camera device 10, so they will not be explained in detail here.

[0059] In addition, after activating the multiple exposure function, the photographic device 10 and its object detection method can selectively further analyze the second pixel variation between the second reference image I2r and the second detection image I2d to determine whether to adjust the second exposure parameters. Since this invention uses an overexposed image to detect a low-contrast target object Ot, if the second pixel variation (e.g.) is analyzed... Figure 9 If the overexposed area is determined to be insufficient or excessive based on the difference in the second pixel array (D_LS), then the object detection method of the present invention needs to increase or decrease the second exposure parameter accordingly based on the degree of insufficient or excessive overexposed area, so that the object detection method of the present invention has preferred object recognition accuracy.

[0060] Please see Figure 10 , Figure 10 This is a schematic diagram of the brightness histogram distribution information of the first reference image I1r according to an embodiment of the present invention. The previously disclosed embodiments disclose that the second exposure parameters can be obtained using ambient light information or pixel brightness calculation values ​​of the first reference image I1r; however, the present invention can also obtain the second exposure parameters using other methods. For example, in some possible variations, the photographic device 10 and its object detection method can first obtain the brightness histogram distribution information of the first reference image I1r and / or the first detection image I1d, in order to... Figure 10Taking the brightness histogram distribution information of the first reference image I1r as an example, the object detection method divides the brightness histogram distribution information into a high-brightness distribution area A1 and a low-brightness distribution area A2. Then, based on the distribution ratio of the high-brightness distribution area A1 and the low-brightness distribution area A2, an overexposure value is estimated or calculated. Based on this, the overexposure value is regarded as or converted into a second exposure parameter. Regardless of the method used to calculate the second exposure parameter, the purpose is to allow the photographic device 10 to capture an overexposed image as the second reference image I2r and the second detection image I2d. The degree of overexposure of the overexposed image can be optimized by obtaining the preferred second exposure parameter through various embodiments disclosed in this invention.

[0061] Accordingly, the object detection method of the present invention can also analyze the brightness histogram distribution information of the second reference image I2r and / or the second detection image I2d to determine and calculate whether to adjust the second exposure parameter. Its application method is the same as the operation mode of calculating the first exposure parameter using the brightness histogram distribution information of the first reference image I1r and / or the first detection image I1d, and will not be described again here.

[0062] In summary, the photographic device and object detection method of the present invention preferably utilize both short-exposure images (i.e., a first reference image and a first detection image generated with a first exposure parameter) and long-exposure images (i.e., a second reference image and a second detection image generated with a second exposure parameter) simultaneously to determine the presence of a target object. Relying solely on short-exposure images may fail to correctly detect low-contrast target objects in low-light environments; relying solely on long-exposure images may result in oversaturation of certain parts of the image due to the bright characteristics of the target object, making detection and judgment difficult. Therefore, the present invention employs a parallel object detection mode using both short-exposure and long-exposure images, and simultaneously provides an automatic off and on function for multiple exposures. Under specific conditions, either short-exposure or long-exposure image object detection modes can be selectively used to balance recognition accuracy and power saving efficiency.

[0063] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An object detection method that can effectively improve the accuracy of identification, characterized in that, The object detection method includes: A first reference image and a second reference image covering a specific monitoring area are obtained during non-detection periods using a first exposure parameter and a second exposure parameter greater than the first exposure parameter. Using the first exposure parameter and the second exposure parameter, a first detection image and a second detection image covering the specific monitoring area are obtained respectively during object detection; Calculate the first pixel variation between the first reference image and the first detected image, and the second pixel variation between the second reference image and the second detected image; and Analyze the first pixel change and the second pixel change to determine whether the target object is located within the specific monitoring area; The object detection method also includes: If the target object determined by the first pixel change amount matches the target object determined by the second pixel change amount, stop using the second exposure parameter to capture related reference images and detection images.

2. The object detection method as described in claim 1, characterized in that, The object detection method also includes: A region of interest is set within the first reference image and the first detection image to calculate the change in the first pixel within the region of interest.

3. The object detection method as described in claim 1, characterized in that, The object detection method also includes: The first range containing the target object and the second range outside the target object are divided from the first pixel variation. The difference between the first range and the second range is compared to a gradient threshold; and If the difference exceeds the gradient threshold, stop using the second exposure parameter to capture relevant reference and detection images.

4. The object detection method as described in claim 1, characterized in that, The object detection method also includes: Actuate the light detector to obtain ambient light information; and The ambient light information is analyzed to determine whether to extract the second exposure parameters and capture the second reference image and the second detection image.

5. The object detection method as described in claim 4, characterized in that, The object detection method also includes: The brightness information of the ambient light is analyzed to calculate the second exposure parameter accordingly.

6. The object detection method as described in claim 1, characterized in that, The object detection method also includes: The calculated brightness value of the first reference image is compared to a brightness threshold; and If the calculated brightness value is lower than the brightness threshold, the second exposure parameter is extracted to capture the second reference image and the second detection image.

7. The object detection method as described in claim 6, characterized in that, The object detection method also includes: Obtain one of the average brightness value and the maximum brightness value of the first reference image; and The calculated value is transformed using a preset weight to calculate the second exposure parameter accordingly.

8. The object detection method as described in claim 6, characterized in that, The object detection method also includes: Obtain the brightness histogram distribution information of the first reference image; The brightness histogram distribution information is divided into high-brightness distribution areas and low-brightness distribution areas; and The second exposure parameter is calculated based on the distribution ratio of the high-brightness distribution area and the low-brightness distribution area.

9. The object detection method as described in claim 1, characterized in that, The object detection method also includes: Analyze the change in the second pixel to determine whether to adjust the second exposure parameter.

10. A photographic device with high recognition accuracy, characterized in that, The photographic equipment includes: An image acquisition device acquires a first reference image and a second reference image covering a specific monitoring area during non-detection periods using a first exposure parameter and a second exposure parameter greater than the first exposure parameter; and acquires a first detection image and a second detection image covering the specific monitoring area during object detection periods using the first exposure parameter and the second exposure parameter; and The processing unit is electrically connected to the image acquisition unit. The processing unit calculates the first pixel change between the first reference image and the first detection image, and the second pixel change between the second reference image and the second detection image, and analyzes the first pixel change and the second pixel change to determine whether the target object is located within the specific monitoring area. When the processor determines that the target object identified by the first pixel change matches the target object identified by the second pixel change, it stops using the second exposure parameter to capture relevant reference and detection images.

11. The photographic apparatus as claimed in claim 10, characterized in that, The processing unit sets a region of interest within the first reference image and the first detected image to calculate the change in the first pixel within the region of interest.

12. The photographic apparatus as claimed in claim 10, characterized in that, The processor divides the target object into a first range and a second range outside the target object from the first pixel variation, compares the difference between the first range and the second range with a gradient threshold, and stops using the second exposure parameter to capture relevant reference and detection images when the difference exceeds the gradient threshold.

13. The photographic apparatus as claimed in claim 10, characterized in that, The photographic device further includes a light detector electrically connected to the processing unit to acquire ambient light information. The processing unit analyzes the ambient light information to determine whether to extract the second exposure parameters and capture the second reference image and the second detection image.

14. The photographic apparatus as claimed in claim 13, characterized in that, The processor analyzes the brightness information of the ambient light to calculate the second exposure parameter accordingly.

15. The photographic apparatus as claimed in claim 10, characterized in that, The processor compares the calculated brightness value of the first reference image with a brightness threshold, and extracts the second exposure parameter when the calculated brightness value is lower than the brightness threshold to capture the second reference image and the second detection image.

16. The photographic apparatus as claimed in claim 15, characterized in that, The processor obtains one of the average brightness value and the maximum brightness value of the first reference image, and uses a preset weight to transform the calculated value to calculate the second exposure parameter accordingly.

17. The photographic apparatus as claimed in claim 15, characterized in that, The processor obtains the brightness histogram distribution information of the first reference image, divides the brightness histogram distribution information into a high brightness distribution area and a low brightness distribution area, and calculates the second exposure parameter accordingly using the distribution ratio of the high brightness distribution area and the low brightness distribution area.

18. The photographic apparatus as claimed in claim 10, characterized in that, The processor analyzes the change in the second pixel to determine whether to adjust the second exposure parameter.

Images