An Overview of Image Generation of Industrial Surface Defects.

Intelligent defect detection technology combined with deep learning has gained widespread attention in recent years. However, the small number, and diverse and random nature, of defects on industrial surfaces pose a significant challenge to deep learning-based methods. Generating defect images can effectively solve this problem. This paper investigates and summarises traditional defect generation and deep learning-based methods. It analyses the various advantages and disadvantages of these methods and establishes a benchmark through classical adversarial networks and diffusion models. The performance of these methods in generating defect images is analysed through various indices. This paper discusses the existing methods, highlights the shortcomings and challenges in the field of defect image generation, and proposes future research directions. Finally, the paper concludes with a summary.

Improved imaging of extremely-slight transparent aesthetic defects using a saturation level-guided method.

The quality-control process of polarizer production is hampered by the presence of extremely-slight transparent aesthetic defects (ESTADs). The saturated imaging method based on stripe structured backlight can effectively improve the imaging contrast of ESTADs. However, the contrast is very sensitive to the saturation degree, which requires careful manual selection. This paper presents a saturation level-guided image enhancement method that is simple to deploy in industrial settings. First, a new definition of the saturation level for structured backlit imaging with translation, scale, and rotation invariance is proposed. Then, an empirical model of contrast versus saturation level is established. Using the contrast data measured at five saturation levels, the optimal saturation level can be estimated using the parameter optimization method. The experimental results demonstrate that the method is effective, easy to use, and an improvement of imaging effects for transparent thin-film defect detection algorithms.

In-line flat-top comb filter based on a cascaded all-solid photonic bandgap fiber intermodal interferometer.

In this paper, an in-line comb filter with flat-top spectral response is proposed and constructed based on a cascaded all-solid photonic bandgap fiber modal interferometer. It consists of two short pieces of all-solid photonic bandgap fiber and two standard single-mode fibers as lead fibers with core-offset splices between them. The theoretical and experimental results demonstrated that by employing a cut and resplice process on the central position of all-solid photonic bandgap fiber, the interference spectra are well tailored and flat-top spectral profiles could be realized by the controllable offset amount of the resplice. The channel position also could be tuned by applying longitudinal torsion with up to 4 nm tuning range. Such a flat-top fiber comb filter is easy-to-fabricate and with a designable passband width and flat-top profile.

Mode-beating-enabled stopband narrowing in all-solid photonic bandgap fiber and sensing applications.

In this paper, core-cladding modal beating in a short piece of all-solid photonic bandgap fiber (AS-PBF) is observed in longitudinal propagation direction. It is demonstrated that at the stopband range of AS-PBF, the power could transfer back and forth between the fiber core and the first layer of high-index rods. Both experimental results and the theoretical analysis from transverse coupled mode theory confirm that the 3-dB width of the sharp stopband could be significantly narrowed by multicycles of such core-cladding modal couplings, which is of great benefit to the high-resolution sensing applications. Based on such a guiding regime, a high-temperature sensor head is also made and its response to temperature is tested to be of 59.9 pm/°C.

Sensitivity-enhanced high-temperature sensing using all-solid photonic bandgap fiber modal interference.

A wavelength-encoded interferometric high-temperature sensor based on an all-solid photonic bandgap fiber (AS-PBF) is reported. It consists of a small piece of AS-PBF spliced core offset with standard single-mode fibers. Two core modes LP(01) and LP(11) are conveniently utilized as optical arms to form Mach-Zehnder-type interference at both the first and the second photonic bandgaps, and the maximum extinction ratio exceeds 25 dB. Experimental and theoretical investigation of its response to temperature confirms that high temperatures up to 700 °C can be effectively sensed using such an AS-PBF interferometer, and benefiting from a large effective thermo-optic coefficient of fiber structure, the sensitivity can be significantly enhanced (71.5 pm/°C at 600 °C).

Fast and Accurate Light Field View Synthesis by Optimizing Input View Selection.

There is a trade-off between spatial resolution and angular resolution limits in light field applications; various targeted algorithms have been proposed to enhance angular resolution while ensuring high spatial resolution simultaneously, which is also called view synthesis. Among them, depth estimation-based methods can use only four corner views to reconstruct a novel view at an arbitrary location. However, depth estimation is a time-consuming process, and the quality of the reconstructed novel view is not only related to the number of the input views, but also the location of the input views. In this paper, we explore the relationship between different input view selections with the angular super-resolution reconstruction results. Different numbers and positions of input views are selected to compare the speed of super-resolution reconstruction and the quality of novel views. Experimental results show that the speed of the algorithm decreases with the increase of the input views for each novel view, and the quality of the novel view decreases with the increase of the distance from the input views. After comparison using two input views in the same line to reconstruct the novel views between them, fast and accurate light field view synthesis is achieved.

Some features of the photonic crystal fiber temperature sensor with liquid ethanol filling.

We introduce a novel photonic crystal fiber (PCF) temperature sensor that is based on intensity modulation and liquid ethanol filling of air holes with index-guiding PCF. The mode field, the effective refractive index and the confinement loss of PCF were all found to become highly temperature-dependent when the thermo-optic coefficient of the liquid ethanol used is higher than that of silicon dioxide and this temperature dependence is an increasing function of the d/Lambda ratio and the input wavelength. All the experiments and simulations are discussed in this paper and the temperature sensitivity of transmission power was experimentally determined to be 0.315 dB/ degrees C for a 10-cm long PCF.

Saturated Imaging for Inspecting Transparent Aesthetic Defects in a Polymeric Polarizer with Black and White Stripes.

Machine vision systems have been widely used in industrial production lines because of their automation and contactless inspection mode. In polymeric polarizers, extremely slight transparent aesthetic defects are difficult to detect and characterize through conventional illumination. To inspect such defects rapidly and accurately, a saturated imaging technique was proposed, which innovatively uses the characteristics of saturated light in imaging by adjusting the light intensity, exposure time, and camera gain. An optical model of defect was established to explain the theory by simulation. Based on the optimum experimental conditions, active two-step scanning was conducted to demonstrate the feasibility of this detection scheme, and the proposed method was found to be efficient for real-time and in situ inspection of defects in polymer films and products.