An uniform illumination system based on Laser3D imaging is designed and the validity of the system is proved by taking simulation experiments. 64 paths of laser beam with average energy are emitted after the narrow infra-red laser pulse emitted by fiber laser device passing through the fiber splitter. The exit array is placed around the focus spot of the optical system to produce dot-array lightening. The laser beam is reflected by target incidents on the receiving system, and then images in APD detector array, where each detection pixel corresponds to each point light source. The system designed can uniformize the light distribution and enhance the utilization efficiency of the laser light.

Key words:

fiber laser device; fiber array; uniform illumination; fiber splitter; planar-array detector

Laser imaging technology is widely used in space exploration and military research. Among them, the uniformity of laser illumination is the decisive factor to improve the imaging accuracy and increase the imaging distance. Therefore, how to improve the problem of insufficient uniformity of light intensity distribution and low energy utilization in laser lighting systems has always been one of the research hotspots.

The traditional laser lighting adopts flood lighting, that is, a laser beam that can cover the entire field of view is emitted at one time. Analysis of the light intensity distribution of the Gaussian beam shows that the divergence angles of the x and y axes (vertical and parallel to the junction plane) are different, and the spot of the collimated laser beam is elliptical, so beam shaping is also required to make the illumination beam regular and uniform. .

The earliest method used to shape the Gaussian beam is the method of combining lenses, which uses 2 to 3 lenses with different surface shapes to change the divergence angles of the Gaussian beam x-axis and y-axis to improve its output shape and the uniformity of the spot.

For example, the use of double-cylindrical mirror combination shaping has the advantages of simple optical system design, convenient installation and easy operation; with the development of the semiconductor industry, the improvement of lithography and microfabrication technology. Since the 1980s, a series of new micro-lens array fabrication technologies have emerged one after another. Therefore, laser illumination methods based on micro-lens arrays have been widely used. The basic idea is to first divide the incident laser into several sub-beams, and then These beamlets are superimposed on the target with a focusing lens.

The splitting of the beam and the superposition of the sub-beams eliminate the influence of the uneven distribution of the incident laser light intensity and achieve the purpose of uniform illumination of the back focal plane of the focusing lens. Uniform illumination, Daman grating is a binary optical element and a binary phase grating with a special aperture function. Light spot array with uniform distribution of light intensity. It avoids the uneven distribution of spectral point light intensity caused by the intensity envelope of the sinc function in general amplitude gratings.

Since the existing laser manufacturing technology is not mature enough, the output Gaussian beam illumination effect is poor, so the above-mentioned related illumination methods using lens group shaping have the problems of not uniform light intensity distribution on the illumination surface and low energy utilization rate.

However, the high cost and difficult procurement of Dammann gratings make it more difficult to study its application in the field of laser lighting.

Therefore, this paper proposes a lattice illumination method based on fiber arrays, which uses fiber lasers, fiber splitters and fiber splicing technology to achieve uniform light intensity illumination. The advantage is that uniform illumination can be achieved. The design can realize the co-optical axis of the transmitting optical system and the receiving optical system, which improves the problems that the light intensity distribution of the traditional laser lighting system is not uniform and the energy utilization rate is low, and the system also increases the detection distance of the system.

(1)Types and selection of beam splitters

According to the different modes of transmission points, optical fibers can be divided into single-mode fibers and multi-mode fibers. The so-called ‘mode’ refers to the angle of the light beam entering the fiber.

The core diameter of the multimode fiber is 50-62.5 μm, the core diameter of the single-mode fiber with a cladding outer diameter of 125 μm is 8.3 μm, and the outer cladding diameter is 125 μm.

The working wavelength of the optical fiber has a short wavelength of 0.85 μm, a long wavelength of 1.31 μm and 1.55 μm.

The fiber loss generally decreases with the wavelength. The multi-mode fiber has a thick core, low transmission speed, short distance, and poor overall transmission performance, but its cost is relatively low; single-mode fiber can only allow one beam of light to propagate, so the single-mode fiber has no mode dispersion characteristics, so the core of the single-mode fiber is relatively thin, the transmission frequency bandwidth and capacity are large, and the transmission distance is long.

Due to the relatively difficult coupling between single-mode fiber and optical devices, high-power single-mode fiber lasers have technical defects in design, which are very rare in the market.

In this experiment, a 915nm pulsed multimode fiber laser is used, and the output is 62.5/125μm multimode fiber, and the power is 20w.

Optical fiber splitter is an optical device that realizes the distribution or combination of optical signal power between different fibers. It is formed by the mutual exchange of guided wave energy in the core area of different optical fibers adjacent to the optical fiber.

According to the different light splitting principles, fiber splitters can be divided into waveguide type and fused taper type. The multimode fiber splitter is a fused taper type with working windows of 850nm, 1300nm and 1510nm. The 850nm splitter has a good splitting effect on the laser used in the experiment, and its loss is within the allowable range.

(2) Spectral loss

Optical fiber splitter has loss, and its loss sources include: splitter loss, plug loss, and transmission loss. The loss caused by the increase of the fiber length belongs to the transmission loss of about 0.2 ~ 0.35dB/km. Since the length of the fiber used in the experiment is very short, this part of the loss can be ignored; the plug loss is the loss generated when the laser and the optical splitter are coupled, usually note 0.3dB/piece; Return loss refers to the ability of the connector to suppress the reflection of the optical power of the link. The return loss of optical splitters on the market is fixed; the largest loss in the splitting process comes from the splitter itself, and the splitting loss and additional loss can be expressed as

When calculating the splitting loss,
a is the output power of a certain channel, and
b is the total power of the output;

when calculating the additional loss,
a is the total power of each output, and
b is the input power.

Knowing the power of the light source and the additional loss of the optical splitter in this paper, the power value at the output end of the optical splitter can be calculated according to the above formula. For example, if a 1:16 optical splitter is used, the additional loss is 0.6dB, and the input power is 20w, then there is 0.6=-10lg (a/20), and a=15.17w, a represents the total power of each branch at the output end, so The output power of each channel is about Pi=0.95w.

As shown in Figure 1, it is the structure diagram of the lighting system.

Fig. 1 System Optical Path

The fiber pulse laser emits a narrow-pulse infrared laser beam. The laser beam passes through the fiber splitter, and one laser is evenly divided into 64 lasers according to the energy. When the array exits, the end face of the fiber array becomes an 8×8 lattice pulsed laser light source. The divergent laser beam changes the divergence angle and becomes a contracted laser beam. The propagation direction of the laser beam is rotated 90 degrees by the plane mirror and then imaged on the target. Above, the laser beam reflected by the target is received by the receiving optical system, and then imaged on the APD area array detector.

As shown in Figure 2, bare fibers with an outer diameter of 125 μm are closely arranged in a square 8×8 format, and the core spacing is 125 μm to form an optical fiber array.

Fig. 2 Schematic diagram of fiber end face

As the outgoing end face of the system, the other end of each fiber is connected to a fiber optic splitter with a standard FT/APC connector.

In order to realize that the area array detector is completely covered by the laser illumination beam, the core spacing between adjacent fibers in the fiber array, the center spacing between adjacent pixels in the area array detector, and the focal lengths of the transmitting and receiving optical systems must be Satisfy certain algebraic conditions. In this system, there are two object-image relationships, the light-emitting point on the fiber array and the illuminated point on the target are a set of object-image relationships, expressed as

The illuminated point on the target is different from the pixel point on the area array detector. The outer set of object-image relationship can be expressed as

where l_t is the distance from the optical fiber array to the transmitting optical system, l_r is the distance between the receiving optical system and the area array detector; L is the detection distance; ƒ_t and ƒ_r are the focal lengths of the transmitting and receiving optical systems, respectively. Then the total vertical magnification of this system is

Because the system is mainly used in long-distance imaging, there is L»l_r, and l_r≈ƒ_r, the above formula changes as

Therefore, the system must satisfy the following formula

where h_s is the center spacing between adjacent pixels in the area array detector; h_f is the fiber core spacing between adjacent fibers in the fiber array.

In the process of emitting the laser beam in the fiber array to the transmitting optical system, in order to pursue the maximum energy utilization rate, the numerical aperture of the transmitting optical system must be larger than the numerical aperture of the optical fiber, that is, the laser light emitted from the optical fiber is within a certain divergence angle range. The diameter of the emission optical system needs to be large enough so that all the outgoing laser light can pass through the emission optical system.

The distance equation of lidar is the theoretical basis of the principle of lidar ranging, which theoretically analyzes the echo power and the maximum detectable distance of the target reflected by the laser ranging system. When the laser detection system is determined, the transmittance of the atmosphere to the laser beam and the reflection characteristics of the detection target surface are the main factors that affect the received echo power of the lidar. The lidar distance equation determines the detection performance of lidar, which is expressed as

where
τ_a is the loss factor of laser energy in atmospheric propagation;
τ₀ is the combined factor of the transfer efficiency of the receiving optical system and the quantum efficiency of the detector.

The energy of the pulse emitted by the laser has a time dependence, and its energy varies with time. The mixed model composed of the Gaussian model and the negative parabolic model is used to describe the laser pulse.

The target contour parameter is represented by T_p(t_kk), t_kk represents the kth receiver time sampling, and T_p(t_kk) represents the product of the surface area and reflectivity of the surface where the corresponding target distance is located at the sampling time point of t_kk. The sum of the products of the echo pulse waveform of each sampling point and the corresponding target contour is the total power detected by the receiver

The noise of the system usually includes speckle noise, background noise and thermal noise. The speckle noise can be simulated as a negative binomial free variable to express the number of detected photons, its mean is equal to the number of photons of the laser signal, and the variance is equal to

where
N is the number of photons of the laser signal;
M is the degree of freedom of the laser, its value is 1 for perfectly coherent light, and its value is infinite for perfectly incoherent light.

When the degree of freedom of the laser is smaller, that is, the coherence of the laser is higher, the laser speckle variance is larger. The figure below is the photon echo image with speckle noise when the laser with a degree of freedom of 1 illuminates a single-surface target. Due to the Poisson nature of background noise, its statistical characteristics obey a Poisson distribution whose parameter is the expected number of photons, and the expected number of photons is

In the formula,
S_IB is the background light intensity;
Δλ is the bandwidth of the optical bandpass filter of the detector, which is usually used to reduce the background radiation, and the bandwidth is usually nm and below;
A_B is the area of the target surface;
h is the common Ranke constant;
v is the frequency of light, and
N is the number of photons generated by the dark current of the photodetector.

The simulation experiment uses the common lens group to shape the Gaussian beam for the conventional system of floodlighting as the comparison of the simulation. Since the simulated pulse energy represents the energy of a single pixel, the conventional system can choose two receivers with different filling factors (photosensitive effective area/total pixel area = filling factor, FillFactor). Here, the filling factor is selected as 16: 25 and 9:25 CCD. As shown in Fig. 3, the simulation results of the three systems for the illumination of the plane target.

Fig. 3 Planar target echo

The simulation results after adding noise and contour factors are shown in Figure 4, which are the illumination echo images of the three systems to the inclined plane target, and the performance differences of the three systems can be seen intuitively.

Fig. 4 Comparison of slope target waveforms

The echoes of the three systems for the target illumination on the step surface are shown in Fig. 5.

Fig. 5 Comparison of target waveforms on stepped surface

The settings of the parameters in the simulation experiment are shown in the table below.

Table 1 Simulation parameters

In the simulation test, the laser beam pulse width, laser beam divergence angle, laser beam coherence degree, detector pixel size, receiving system aperture, receiving optical system focal length and temperature are selected according to the actual laboratory equipment; the laser beam peak power is selected according to the detection distance to make a selection. Other parameters are selected according to the parameters in related papers.

In the actual test, there are certain differences between the test parameters and the parameters used in the simulation experiment, specifically:
(1) The waveform of the laser beam emitted in the simulation is ideally designed, but such an ideal waveform cannot be achieved in practice;
(2) The aberration of the transmitting and receiving optical system is not considered in the simulation, which needs to be considered in practice;
(3) The influence of atmospheric turbulence on the laser echo waveform is not considered in the simulation, which needs to be considered in practice;
(4) The influence of the fiber on the laser echo waveform is not considered in the simulation, which needs to be considered in practice.

Based on the above simulation comparison, the target contour and system noise will have an important impact on the energy distribution of the echo signal of the lidar system, and under the conventional system, this effect is especially obvious. When the target surface shape is not a simple vertical plane, due to the waveform Due to the widening of the single point and the influence of system noise, the echo signal energy at a single point is attenuated to a certain extent, and it is easily overwhelmed by the noise. However, the system discussed in this paper, because the single point energy is concentrated and is not affected by the surface shape of the target object, maintains the Very good transmission and detection stability.