In this work, performance improvements are described for a low-power consumption non-dispersive infrared (NDIR) methane (CH₄) gas sensor using customised optical thin film bandpass filters (BPFs) centered at 3300 nm. BPFs shape the spectral characteristics of the combined mid-infrared III–V based light emitting diode (LED)/photodiode (PD) light source/detector optopair, enhancing the NDIR CH₄ sensor performance. The BPFs, deposited using a novel microwave plasma-assisted pulsed DC sputter deposition process, provide room temperature deposition directly onto the temperature-sensitive PD heterostructure. BPFs comprise germanium (Ge) and niobium pentoxide (Nb₂O₅) alternating high and low refractive index layers, respectively. Two different optical filter designs are progressed with BPF bandwidths (BWs) of 160 and 300 nm. A comparison of the modelled and measured NDIR sensor performance is described, highlighting the maximised signal-to-noise ratio (SNR) and the minimised cross-talk performance benefits. The BPF spectral stability for various environmental temperature and humidity conditions is demonstrated.

3D object detection based on point cloud data in the unmanned driving scene has always been a research hotspot in unmanned driving sensing technology. With the development and maturity of deep neural networks technology, the method of using neural network to detect three-dimensional object target begins to show great advantages. The experimental results show that the mismatch between anchor and training samples would affect the detection accuracy, but it has not been well solved. The contributions of this paper are as follows. For the first time, deformable convolution is introduced into the point cloud object detection network, which enhances the adaptability of the network to vehicles with different directions and shapes. Secondly, a new generation method of anchor in RPN is proposed, which can effectively prevent the mismatching between the anchor and ground truth and remove the angle classification loss in the loss function. Compared with the state-of-the-art method, the AP and AOS of the detection results are improved.

The authors wish to modify the funding section as follows to this paper