Search

Edge detection is one of the fundamental computer vision tasks. Recent methods for edge detection based on a convolutional neural network (CNN) typically employ the weighted cross-entropy loss. Their predicted results being thick and needing post-processing before calculating the optimal dataset scale (ODS) F-measure for evaluation. To achieve end-to-end training, we propose a non-maximum suppression layer (NMS) to obtain sharp boundaries without the need for post-processing. The ODS F-measure can be calculated based on these sharp boundaries. So, the ODS F-measure loss function is proposed to train the network. Besides, we propose an adaptive multi-level feature pyramid network (AFPN) to better fuse different levels of features. Furthermore, to enrich multi-scale features learned by AFPN, we introduce a pyramid context module (PCM) that includes dilated convolution to extract multi-scale features. Experimental results indicate that the proposed AFPN achieves state-of-the-art performance on the BSDS500 dataset (ODS F-score of 0.837) and the NYUDv2 dataset (ODS F-score of 0.780). [ABSTRACT FROM AUTHOR]

The present work investigated a solar assisted heat pump system for drying Chinese wolfberry. The kinetic characteristic was firstly analyzed through a series of lab experiments. It was concluded that the Page model was the most suitable for predicting the heat and mass transfer of the wolfberry. Based on the wolfberry kinetic model, solar collector model and chamber air model, the coupled drying system model was developed. The accuracy of the mathematic model was determined through comparing with the preliminary experimental results. The influence of operating conditions on the thermal and energy performance of the dryer for the different operating mode was discussed. The drying weight of no more than 75 kg may be preferable in the stand-alone solar drying mode, and less than 15 h was needed to be dried. The electric energy consumption in the solar assisted the heat pump drying mode was lower than that in the stand-alone heat pump mode, and it was recommended that about 50 kg of wolfberry to be dried in the solar assisted heat pump system. Compared to the autumn drying, the reduction in the electric energy consumption was around 9.1 kWh during the 11 h summer drying process. The obtained results demonstrated the feasibility of the combined system for drying wolfberry, and also can provide the basic theoretical and experimental data support for the following research. [ABSTRACT FROM AUTHOR]

In recent years, convolutional neural networks (CNN) have been widely used in image denoising for their high performance. One difficulty in applying the CNN to medical image denoising such as speckle reduction in the optical coherence tomography (OCT) image is that a large amount of high-quality data is required for training, which is an inherent limitation for OCT despeckling. Recently, deep image prior (DIP) networks have been proposed for image restoration without pre-training since the CNN structures have the intrinsic ability to capture the low-level statistics of a single image. However, the DIP has difficulty finding a good balance between maintaining details and suppressing speckle noise. Inspired by DIP, in this paper, a sorted non-local statics which measures the signal autocorrelation in the differences between the constructed image and the input image is proposed for OCT image restoration. By adding the sorted non-local statics as a regularization loss in the DIP learning, more low-level image statistics are captured by CNN networks in the process of OCT image restoration. The experimental results demonstrate the superior performance of the proposed method over other state-of-the-art despeckling methods, in terms of objective metrics and visual quality. [ABSTRACT FROM AUTHOR]

In order to better release the heat generated by the electronic components, a novel heat dissipation system is proposed, which combines a microchannel heat pipe (MHP) with a high thermal conductivity and a radiative plate with a high emissivity at nighttime. First, a simple testing rig was made with an MHP and a radiative plate, where the radiative plate was made of acrylic resin, a curing agent, thinner, and aluminum plate, and had strong radiative cooling at nighttime. Second, the mathematical model was initially established and verified using experiments, where it was shown that the agreement between numerical and experimental data was well within experimental uncertainties. Comprehensive simulation investigations were conducted by varying wind speed, relative humidity, the cloudiness coefficient, dimension of the radiative plate, and tilted angle. The results show that: (1) the emissivity of the radiative plate was about 0.311 in the daytime and about 0.908 in the nighttime; (2) the influence of wind speed on reducing the component surface temperature was greater than the cloudiness coefficient and relative humidity; (3) the width of the radiative plate had a greater effect on heat dissipation than on its length, and the maximum size of radiative plate was recommended to be 400 mm × 400–500 mm (length × width), which was equipped with a single MHP (width: 60 mm). Additionally, the tilted angle of the radiative plate should be kept within 30° of the horizontal level. In conclusion, the novel heat dissipation system had a superior application value for providing assisted electronic component cooling in the nighttime. [ABSTRACT FROM AUTHOR]

This paper presents a fast algorithm for texture-less object recognition, which is designed to be robust to cluttered backgrounds and small transformations. At its core, the proposed method demonstrates a two-stage template-based procedure using an orientation compressing map and discriminative regional weight (OCM-DRW) to effectively detect texture-less objects. In the first stage, the proposed method quantizes and compresses all the orientations in a neighborhood to obtain the orientation compressing map which then is used to generate a set of possible object locations. To recognize the object in these possible object locations, the second stage computes the similarity of each possible object location with the learned template by using discriminative regional weight, which can effectively distinguish different categories of objects with similar parts. Experiments on publiclyavailable, texture-less object datasets indicate that apart from yielding efficient computational performance, the proposed method also attained remarkable recognition rates surpassing recent state-of-the-art texture-less object detectors in the presence of high-clutter, occlusion and scale-rotation changes. It improves the accuracy and speed by 8% and 370% respectively, relative to the previous best result on D-Textureless dataset. [ABSTRACT FROM AUTHOR]