In recent years, AI technology using Neural Network (NN) has made remarkable progress and is used for highly accurate classification, object detection, and anomaly detection in sensing. The difficulties with high-accuracy NN are the long processing time and high-power consumption. As one solution, an optical neural network (ONN), which realizes NN by diffraction and propagation of light, has attracted attention as an implementation method with ultra-high speed and low power consumption. Although many of the prior studies on ONN are related to classification, ONN has the potential to be applied to various tasks. As one example, the use of ONN has the possibility of ultra-fast object detection. In this study, simulations and experiments were conducted to verify the possibility of detection by ONN. Metal nuts were selected as the detection targets as a representative example of mass-produced industrial parts. In the experiment, SLM was used to implement the data input layer as phase input and the trained diffraction layer. First, the case of a single detection target in the input data was demonstrated. The precision for the 551-input data was 96.4 % in the experiment. In the data that could be detected correctly, the root mean square error between the inferred and correct positions was 2.2 % of the metal nut size. Next, another experiment has confirmed that ONN can detect multiple targets accurately. In addition, we examined ONN that uses light transmitted through the sample and found that the inference process finished within 4.17 msec (the response time of the CMOS of this setup). The results show that ONN can accurately and rapidly detect objects.
In recent years, sensing and imaging have significantly progressed due to AI algorithms such as Neural Network (NN). The main issues of applying NNs to information processing are the limited processing speed and high energy consumption of electronic processors. Optical Neural Network (ONN), which utilizes diffraction and propagation of light for processing, is an intriguing implementation of an ultra-fast and low-energy-consuming NN. However, previous studies of ONN are mainly on simulations due to the experimental difficulty of processing more than hundreds of input data. In hardware implementations, the performance or the classification accuracy of ONNs can be reduced by the noise and the displacements. Therefore, not only must the ONN achieve high theoretical accuracy, but it must also be robust to these experimental errors. In this study, the classification of 1,000 MNIST input data (100 data for each of 10 classes) was realized experimentally as well as theoretically, taking advantage of our novel setup with a variable spatial light modulator (SLM). With our experimental configuration, we investigated the classification accuracy with several loss functions for the ONN training. The inference accuracy of the MNIST classification task was up to 97% in the simulation and ~95% in the experiment by softmax-cross-entropy (SCE) loss function. Also, the classification accuracy of 98% for a Surface crack classification and 93% for a Pollen classification was achieved experimentally. These results show that SCE can realize high-accuracy classification in the ONN implementation. Our results revealed the high application capability of the optical neural network for practical sensing tasks.
A high power single-mode fiber laser has received a lot of attention in various materials processing fields. In order to carry out processing under proper conditions with single-mode fiber laser, the dynamic flows of fluids inside the materials should be precisely understood. We carried out the numerical analysis of materials processing using a 5-kW single-mode fiber laser. The dynamic flows of the copper fluid during the bead-on-plate tests are calculated and the results agree well with the experimental results. The processing with the different scan speed and spot size, and at the output power of 10 kW are numerically calculated as well. The proper processing conditions with less spatters and dross for each materials can be determined by the dynamic analysis.
Stimulated Raman scattering gives back reflection sensitivity to a high power fiber laser. Therefor SRS suppression is necessary in order to realize stable laser processing by a high power fiber laser. A 5-kW single-mode ytterbium doped fiber laser with a 20-m long delivery fiber has been realized. The fiber laser is an all-fiber single-stage Fabry-Perot system in a co-pumping configuration. The optical to optical efficiency was 80% at the output power of 5.0 kW. And the M-squared figure of 1.3 was obtained. The Stokes light by SRS is suppressed to 45 dB below the laser output by using fibers with the effective mode area of 600 μm2 . While SRS was well suppressed, four wave mixing was observed with the frequency shift of ~6 THz. Four wave mixing between the fundamental mode and the secondary modes is believed to take place. Four wave mixing is believed not to give back reflection sensitivity to the fiber laser. The 5-kW single-mode fiber laser was applied to laser processing. Bead-on-plate tests were carried out with a galvanometer scanner. The laser ran without stopping nor damaging the laser system even during processing highly reflective material. This implies that our SRS suppressed single-mode fiber laser can be used practically in most of processing systems.
A 3 kW single stage all-fiber Yb-doped single-mode fiber laser with bi-directional pumping configuration has been demonstrated. Our newly developed high-power LD modules are employed for a high available pump power of 4.9 kW. The length of the delivery fiber is 20 m which is long enough to be used in most of laser processing machines. An output power of 3 kW was achieved at a pump power of 4.23 kW. The slope efficiency was 70%. SRS was able to be suppressed at the same output power by increasing ratio of backward pump power. The SRS level was improved by 5dB when 57% backward pump ratio was adopted compared with the case of 50%. SRS was 35dB below the laser power at the output power of 3 kW even with a 20-m delivery fiber. The M-squared factor was 1.3. Single-mode beam quality was obtained. To evaluate practical utility of the 3 kW single-mode fiber laser, a Bead-on-Plate (BoP) test onto a pure copper plate was executed. The BoP test onto a copper plate was made without stopping or damaging the laser system. That indicates our high power single-mode fiber lasers can be used practically in processing of materials with high reflectivity and high thermal conductivity.
A 2 kw single-mode fiber laser with a 20-m long delivery fiber and high back reflection resistance has been demonstrated. An Yb-doped fiber with large core size and differential modal gain is used to realize high SRS suppression and single-mode operation simultaneously. The 20 m-long delivery fiber gives flexibility to the design of processing systems. An output power of 2 kW is achieved at a pump power of 2.86 kW. The slope efficiency is 70%. The power of the Stokes light is less than -50 dB below the laser power at the output power of 2 kW even with a 20-m delivery fiber. Nearly diffraction-limited beam quality is also confirmed (M2 = 1.2). An output power of 3 kW is believed to be achieved by increasing pumping power. The back reflection resistance properties of the fabricated singlemode fiber laser is evaluated numerically by the SRS gain calculated from measured laser output spectra and fiber characteristics. The acceptable power of the back reflection light into the fiber core is estimated to be 500 W which is high enough for processing of highly reflective materials. The output power fluctuation caused by SRS and back reflection in materials processing will be well suppressed. Our high power single-mode fiber lasers can provide high quality and stable processing of highly reflective materials.
A long-range and high-resolution reflectometry is proposed by synthesis of optical coherence function at region beyond the coherence length has been proposed. We discuss and simulate the principle of the reflectometry system. In basic experiments, the reflectivity distributions are successfully measured at the region beyond the coherence length with a spatial resolution of 19 cm and a measurement range of 1 km.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.