A non-sensing laser active detection system for micro-peek devices

Abstract

The present invention belongs to the technical field of detecting micro-spying devices, and specifically relates to a non-sensing laser active detection system for micro-spying devices, comprising a transmitting system, a receiving system, and a positioning and identification system. The transmitting system's optical path is directed toward a target object, and the transmitting system illuminates a receiving system in the reflected optical path of the target object. The receiving system is electrically connected to the positioning and identification system via a wire. The present invention uses a 1550nm near-infrared laser for active excitation detection. The 1550nm laser has the characteristics of stability and minimal attenuation at close range. Furthermore, it is invisible to the human eye, thus resolving the issues of active detection devices being easily exposed and difficult to conceal. The present invention employs multi-algorithm fusion for auxiliary recognition, forming a highly robust and interference-resistant real-time detection device that supports the simultaneous execution of multiple algorithms to improve recognition rates.

Description

Technical Field

[0001] The present invention belongs to the technical field of detecting micro-peek devices, and in particular relates to a non-sensing laser active detection system for micro-peek devices. Background Art

[0002] Pinhole cameras, micro cameras and other miniature photoelectric eavesdropping devices are currently the mainstream way to steal important information. They also have certain concealment characteristics. The popularity of various tiny devices has also posed a serious threat to the field of information security. Without the use of professional equipment for detection, it is difficult for the human eye to detect them.

[0003] Traditional active detection equipment, driven by cost and portability, uses ordinary light for active detection, leveraging the cat's eye effect to generate echo signals, and then relying on the human eye to subjectively identify the presence of an optoelectronic device. While these advantages include portability and low cost, they also suffer from subjective errors in the human eye, can easily expose the active detection intent, and have low accuracy and recognition rates, making them unsuitable for evidence collection and storage.

[0004] Compared to traditional active detection equipment, the sensorless laser active detection device uses near-infrared lasers, invisible to the human eye. During active detection, the device does not reveal its intention by actively emitting blanket light. The invisible laser perfectly conceals its own activity. Furthermore, the optical receiving system collects echo signals and performs a series of background noise filtering, target feature extraction, and algorithm recognition. This enables high-precision detection and identification, active labeling, alarms, and storage of video evidence of peeping behavior, forming a complete sensorless laser active detection micro-peeping device system.

[0005] Although domestic laser active detection technology started late, research conducted by scholars and teams in the field in recent years has led to rapid development of laser active detection technology using the cat's eye effect. Various methods have been proposed and combined, including circular features of light cross-sections, texture features of echo signal images, and compressed sensing theory. It has also been used in practical applications in anti-sniper systems in the military field.

[0006] In 2016, research into active laser detection and identification technology using 808nm lasers presented the following drawbacks and challenges: Because it is close to the visible spectrum, has a detection range of approximately 5 meters, and uses 1 watt of power, the human eye readily absorbs near-infrared light in this wavelength range during active detection, posing a risk of eye damage. Furthermore, 808nm lasers, being close to the visible light domain, are prone to a "redburst" phenomenon, whereby a faint red light is emitted during active detection, without providing a way to conceal the detection activity.

[0007] In 2020, when using the cat's eye effect to detect and track hidden optoelectronic equipment, blind source separation methods and neural network classification algorithms were used to achieve filtering and identify real echo signals. The final recognition effect can reach more than 90% when there are fewer interferences, but the false alarm rate is higher when the number of interferences increases. Summary of the Invention

[0008] In order to address the technical problems that the above-mentioned active detection equipment does not have the property of non-sensing detection and there is a risk of damage to the human eye, the laser active detection equipment has poor anti-interference ability, low recognition rate, and difficult positioning, and the traditional laser active detection peeping equipment system has a single application environment and strong limitations, the present invention provides a non-sensing laser active detection micro peeping equipment system with a high security threshold, strong anti-interference ability, high recognition rate, and wide applicability.

[0009] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0010] A non-sensing laser active detection system for micro-peek devices includes a transmitting system, a receiving system and a positioning and identification system. A target object is set in the direction of the optical path of the transmitting system, and a receiving system is set on the reflected optical path of the target object irradiated by the transmitting system. The receiving system is electrically connected to the positioning and identification system through a wire.

[0011] The transmitting system includes a laser transmitter and a beam expander, wherein the beam expander is arranged in the optical path direction of the laser transmitter, and a target object is arranged in the optical path direction of the beam expander.

[0012] The receiving system includes an optical lens, a narrowband filter, a focal plane detector and an image sensor. The optical lens is arranged on the reflected light path of the target object, a narrowband filter is arranged in the light path direction of the optical lens, a focal plane detector is arranged in the light path direction of the narrowband filter, the focal plane detector is connected to the image sensor, and the image sensor is connected to the positioning and recognition system through a wire.

[0013] The laser emitter uses a near-infrared laser in the 1550nm band; the spectral response range of the narrow-band filter is 0.9-1.7μm, and the spectral range of the narrow-band filter is 1550±10nm.

[0014] The positioning and identification system is respectively connected to a wired PC terminal, a wireless PC terminal and a wireless display and control APP terminal.

[0015] The positioning and identification system adopts a multi-algorithm fusion auxiliary identification method, which includes a local extreme value watershed algorithm module, a dynamic positioning tracking and identification module, an echo signal feature multiple identification module, and an echo signal image matching algorithm identification module. The positioning and identification system adopts a pan-tilt automatic scanning method.

[0016] A method for processing a non-sensing laser active detection system for a micro-peek device, comprising the following steps:

[0017] S1, collecting echo signal images through image sensors;

[0018] S2, perform median filtering on the echo signal image, using a 3×3 filter window;

[0019] S3. Divide the median filtered data into four parts for parallel processing. The first, second, and third parts of the data are grayscale processed to process the three-channel image into a single-channel grayscale image. The first part of the grayscale processed data enters the local extreme value watershed algorithm module. The second part of the grayscale processed data enters the dynamic positioning tracking and recognition module after binarization processing. The third part of the grayscale processed data enters the echo signal feature multiple recognition module after binarization processing. The fourth part of the data enters the echo signal image matching algorithm recognition module.

[0020] S4. The upper and lower limits of the thresholds are set in the local extreme value watershed algorithm module for the first part of the grayscale processed data, and then the step features are extracted after step-by-step binarization;

[0021] S5, after the second part of the data is binarized, the background frame difference segmentation is used in the dynamic positioning tracking and recognition module to extract motion features;

[0022] S6, after the third part of the data has been binarized, the target image is segmented multiple times in the echo signal feature multiple recognition module to extract morphological features;

[0023] S7. The fourth part of the data is subjected to feature extraction of the echo signal features by using the normalized correlation coefficient matching algorithm in the echo signal image matching algorithm recognition module, and then the echo signal features are matched with suspicion degree;

[0024] S8, integrating and screening multiple features to perform multi-feature recognition of echo signals;

[0025] S9. The recognition result is finally formed based on the feature recognition rate and the matching suspicion degree, completing the multi-algorithm fusion-assisted recognition.

[0026] The method for binarizing the second and third parts of data in S3 is as follows: the pixel values ​​in the grayscale image are 0 to 255, and then binarized to separate them into two colors: 0 and 255, black and white.

[0027] The method of extracting morphological features by segmenting the target image multiple times in S6 is as follows: the extracted morphological feature is roundness C, which is determined by the perimeter and the area. The calculation formula of the roundness C is: The S is the area, and the p is the perimeter.

[0028] The method for extracting the echo signal features by the normalized correlation coefficient matching algorithm in S7 is as follows: the feature extraction formula of the echo signal feature R(x, y) is:

[0029]

[0030] The x and y are the matching image positions, the w and h are the width and height pixel values ​​of the template image, the I(x+x′, y+y′) is the average grayscale value of the matching image, the T(x+x′, y+y′) is the pixel grayscale value of the image to be matched at the x+x′, y+y′ position, and the T(x′, y′) is the pixel grayscale value of the image to be matched at the x′, y′ position.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] This invention employs a cat's-eye effect-based echo signal for active detection. Laser coverage is generated when the laser hits the target. After the receiving system acquires the signal, it uses a combination of median filtering and morphological filtering to remove background noise and interference. Computer vision and image processing algorithms are then used to further identify and annotate the target, addressing the difficulty of detecting indoor micro-spying devices. This invention uses a 1550nm near-infrared laser for active detection. 1550nm lasers offer stability, minimal attenuation at close range, and are invisible to the human eye. This addresses the issues of active detection devices being easily exposed and difficult to conceal. While the algorithms currently used in traditional laser active detection and recognition technologies for indoor applications can maintain high recognition rates, they perform less well in complex environments with increased interference. This invention employs a multi-algorithm fusion-assisted recognition system, creating a robust and interference-resistant real-time detection device that supports the simultaneous execution of multiple algorithms, improving recognition rates. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary, and those skilled in the art can, without inventive effort, derive other implementation drawings based on the provided drawings.

[0034] The structures, proportions, sizes, etc. illustrated in this specification are intended solely to complement the contents disclosed herein and to facilitate understanding and reading by persons skilled in the art. They are not intended to limit the conditions under which the present invention may be implemented and therefore have no substantive technical significance. Any structural modifications, changes in proportions, or adjustments in sizes, without affecting the efficacy and objectives of the present invention, shall remain within the scope of the technical contents disclosed herein.

[0035] Figure 1 It is a structural schematic diagram of the present invention;

[0036] Figure 2 Schematic diagram of the receiving system and echo signal recognition algorithm of the present invention;

[0037] Figure 3 This is a diagram of the software and hardware system architecture of the present invention.

[0038] Among them: 1 is the transmitting system, 2 is the receiving system, 3 is the positioning and identification system, 4 is the target object, 101 is the laser transmitter, 102 is the beam expander, 201 is the optical lens, 202 is the narrowband filter, 203 is the focal plane detector, and 204 is the image sensor. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only part of the embodiments of this application, not all the embodiments. These descriptions are only to further illustrate the features and advantages of the present invention, rather than to limit the claims of the present invention. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0040] The following embodiments of the present invention are described in further detail with reference to the accompanying drawings and examples. The following embodiments are used to illustrate the present invention but are not intended to limit the scope of the present invention.

[0041] In the description of this application, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they can refer to fixed, detachable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in this application based on the specific circumstances.

[0042] In this embodiment, if Figure 1As shown, the distance between the transmitting system 1 and the receiving system 2 and the target object 4 is within 100m, and the effective aperture of the receiving system 2 is within 50mm. The transmitting system 1 and the receiving system 2 are located in the same position. Within the effective distance, the laser emitter 101 of the transmitting system 1 is amplified and scattered by the beam expander 102 to form a light spot sufficient to cover the target optical lens. The impact of laser energy attenuation within this distance is negligible. Unlike conventional cat's eye effect active detection, the laser emitter 101 uses a near-infrared laser in the 1550nm band, forming an invisible detection method that is invisible to the human eye. The receiving system 2 uses a high-precision, high-sensitivity focal plane detector 203 with a spectral response range of 0.9-1.7μm and a narrowband filter 202 of 1550±10nm to effectively detect, receive, and collect data, which is then transmitted to the identification and positioning system 3. Based on the collected data, multiple algorithms are integrated on the embedded ARM platform to assist in the identification and positioning of optoelectronic surveillance devices.

[0043] like Figure 2 As shown, the laser echo signal is detected by the optical receiving system and finally forms a spot image which is transmitted to the algorithm processing and recognition system and is processed in a distributed manner through the following steps:

[0044] Step 1: Perform median filtering on the echo image using a 3×3 filter window.

[0045] Step 2: Data parallel processing, four parts are processed in parallel, and one part of the data is subjected to feature extraction CV_TM_CCOEFF-NORMED by the normalized correlation coefficient matching algorithm. The formula is:

[0046]

[0047] Where x and y are the matching image positions, w and h are the pixel values ​​of the template image's width and height, I(x+x', y+y') is the average grayscale value of the matching image, T(x+x', y+y') is the grayscale value of the pixel at position x+x', y+y' in the image to be matched, and T(x', y') is the grayscale value of the pixel at position x', y' in the image to be matched. The remaining steps process the three-channel image into a single-channel grayscale image, where pixel values ​​range from 0 to 255. This grayscale image is then binarized to produce a black and white image with values ​​of 0 and 255.

[0048] Step 3: Set the upper and lower limits of the threshold in the local extreme watershed algorithm module, and then extract the step feature after gradual binarization; use the background frame difference segmentation to extract the motion feature in the dynamic positioning tracking recognition module; and segment the target image multiple times in the echo signal feature multiple recognition module to extract the morphological feature - roundness C, which is expressed as:

[0049] Determined by the perimeter and area, S is the area and p is the perimeter.

[0050] Step 4: Integrate and filter multiple features to perform multi-feature recognition of echo signals.

[0051] Step 5: The recognition result is finally formed based on the feature recognition rate and the matching suspicion, completing the multi-algorithm fusion assisted recognition.

[0052] like Figure 3 As shown in the figure, the hardware and software system is divided into hardware acquisition layer, data transmission layer, algorithm processing layer, application function layer, and display and control terminal layer from bottom to top. The embedded system platform controls the optical camera and pan-tilt head, and then uses TCP / UDP multi-threaded data transmission and multiple algorithms to complete the identification, tracking and active alarm work, forming a complete set of non-sensing laser active detection micro-peeking equipment system.

[0053] Compared with traditional laser detection active peeping devices, the present invention realizes non-sensing active detection by using the solution of this embodiment to realize non-sensing active detection, which conceals its own detection behavior and greatly reduces the risk of damage to human eyes in the environment.

[0054] This embodiment adopts multi-algorithm fusion auxiliary recognition technology, which increases the overall detection and recognition success rate when a single algorithm has poor robustness and weak anti-interference ability. The traditional algorithm can achieve a recognition rate of 90% when there are fewer interferers, and the false alarm rate exceeds 20% when there are more interferers. In comparison, the algorithm of this embodiment has a recognition rate of 96% when there are fewer interferers, and a false alarm rate of less than 10% when there are more interferers.

[0055] The above only describes in detail the preferred embodiments of the present invention, but the present invention is not limited to the above embodiments. Various changes can be made within the knowledge of ordinary technicians in this field without departing from the purpose of the present invention, and various changes should be included in the scope of protection of the present invention.

Claims

1. A non-sensing laser active detection system for micro-peek devices, characterized by: The invention comprises a transmitting system (1), a receiving system (2) and a positioning identification system (3); a target object (4) is arranged in the direction of the optical path of the transmitting system (1); a receiving system (2) is arranged on the reflected optical path of the target object (4) irradiated by the transmitting system (1); and the receiving system (2) is electrically connected to the positioning identification system (3) via a wire; The positioning and recognition system (3) adopts a multi-algorithm fusion auxiliary recognition method, which includes a local extreme value watershed algorithm module, a dynamic positioning tracking recognition module, an echo signal feature multiple recognition module, and an echo signal image matching algorithm recognition module. The positioning and recognition system (3) adopts a pan-tilt automatic scanning method; The processing method of the non-sensing laser active detection system for micro-peek devices includes the following steps: S1, collecting echo signal images through image sensors; S2, perform median filtering on the echo signal image, using a 3×3 filter window; S3. Divide the median filtered data into four parts for parallel processing. The first, second, and third parts of the data are grayscale processed to process the three-channel image into a single-channel grayscale image. The first part of the grayscale processed data enters the local extreme value watershed algorithm module. The second part of the grayscale processed data enters the dynamic positioning tracking and recognition module after binarization processing. The third part of the grayscale processed data enters the echo signal feature multiple recognition module after binarization processing. The fourth part of the data enters the echo signal image matching algorithm recognition module. S4. The upper and lower limits of the thresholds are set in the local extreme value watershed algorithm module for the first part of the grayscale processed data, and then the step features are extracted after step-by-step binarization; S5, after the second part of the data is binarized, the background frame difference segmentation is used in the dynamic positioning tracking and recognition module to extract motion features; S6, after the third part of the data has been binarized, the target image is segmented multiple times in the echo signal feature multiple recognition module to extract morphological features; S7. The fourth part of the data is subjected to feature extraction of the echo signal features by using the normalized correlation coefficient matching algorithm in the echo signal image matching algorithm recognition module, and then the echo signal features are matched with suspicion degree; S8, integrating and screening multiple features to perform multi-feature recognition of echo signals; S9. The recognition result is finally formed based on the feature recognition rate and the matching suspicion degree, completing the multi-algorithm fusion-assisted recognition.

2. The non-sensing laser active detection system for micro-peek devices according to claim 1, characterized in that: The transmitting system (1) comprises a laser transmitter (101) and a beam expander (102), wherein the beam expander (102) is arranged in the direction of the optical path of the laser transmitter (101), and a target object (4) is arranged in the direction of the optical path of the beam expander (102).

3. The non-sensing laser active detection system for micro-peek devices according to claim 2, characterized in that: The receiving system (2) comprises an optical lens (201), a narrowband filter (202), a focal plane detector (203) and an image sensor (204); the optical lens (201) is arranged on a reflected light path of a target object (4); a narrowband filter (202) is arranged in the light path direction of the optical lens (201); a focal plane detector (203) is arranged in the light path direction of the narrowband filter (202); the focal plane detector (203) is connected to the image sensor (204); and the image sensor (204) is connected to a positioning and recognition system (3) via a wire.

4. The non-sensing laser active detection system for micro-peek devices according to claim 3, characterized in that: The laser emitter (101) uses a near-infrared laser in the 1550 nm band; the spectral response range of the narrow-band filter (202) is 0.9-1.7 μm, and the spectral range of the narrow-band filter (202) is 1550±10 nm.

5. The non-sensing laser active detection system for micro-peek devices according to claim 1, characterized in that: The positioning and identification system (3) is respectively connected to a wired PC terminal, a wireless PC terminal and a wireless display and control APP terminal.

6. The non-sensing laser active detection system for micro-peek devices according to claim 1, characterized in that: The method for binarizing the second and third parts of data in S3 is as follows: the pixel values ​​in the grayscale image are 0 to 255, and then binarized to separate them into two colors: 0 and 255, black and white.

7. The non-sensing laser active detection system for micro-peek devices according to claim 1, characterized in that: The method of extracting morphological features by segmenting the target image multiple times in S6 is as follows: the extracted morphological feature is roundness C, which is determined by the perimeter and the area. The calculation formula of the roundness C is: The S is the area, and the p is the perimeter.

8. The non-sensing laser active detection system for micro-peek devices according to claim 1, characterized in that: The method for extracting the echo signal features by the normalized correlation coefficient matching algorithm in S7 is as follows: the feature extraction formula of the echo signal feature R(x, y) is: The x and y are the matching image positions, the w and h are the width and height pixel values ​​of the template image, the I(x+x′, y+y′) is the average grayscale value of the matching image, the T(x+x′, y+y′) is the pixel grayscale value of the image to be matched at the x+x′, y+y′ position, and the T(x′, y′) is the pixel grayscale value of the image to be matched at the x′, y′ position.

Images