Defect segmentation of apples is an important task in the agriculture industry for quality control and food safety. In this paper, we propose a deep learning approach for the automated segmentation of apple defects using convolutional neural networks (CNNs) based on a U-shaped architecture with skip-connections only within the noise reduction block. An ad-hoc data synthesis technique has been designed to increase the number of samples and at the same time to reduce neural network overfitting. We evaluate our model on a dataset of multi-spectral apple images with pixel-wise annotations for several types of defects. In this paper, we show that our proposal outperforms in terms of segmentation accuracy general-purpose deep learning architectures commonly used for segmentation tasks. From the application point of view, we improve the previous methods for apple defect segmentation. A measure of the computational cost shows that our proposal can be employed in real-time (about 100 frame-per-second on GPU) and in quasi-real-time (about 7/8 frame-per-second on CPU) visual-based apple inspection. To further improve the applicability of the method, we investigate the potential of using only RGB images instead of multi-spectral images as input images. The results prove that the accuracy in this case is almost comparable with the multi-spectral case.
We introduce a no-reference method for the assessment of the quality of videos affected by in-capture distortions due to camera hardware and processing software. The proposed method encodes both quality attributes and semantic content of each video frame by using two Convolutional Neural Networks (CNNs) and then estimates the quality score of the whole video by using a Recurrent Neural Network (RNN), which models the temporal information. The extensive experiments conducted on four benchmark databases (CVD2014, KoNViD-1k, LIVE-Qualcomm, and LIVE-VQC) containing in-capture distortions demonstrate the effectiveness of the proposed method and its ability to generalize in cross-database setup.
Methods for No-Reference Video Quality Assessment (NR-VQA) of consumer-produced video content are largely investigated due to the spread of databases containing videos affected by natural distortions. In this work, we design an effective and efficient method for NR-VQA. The proposed method exploits a novel sampling module capable of selecting a predetermined number of frames from the whole video sequence on which to base the quality assessment. It encodes both the quality attributes and semantic content of video frames using two lightweight Convolutional Neural Networks (CNNs). Then, it estimates the quality score of the entire video using a Support Vector Regressor (SVR). We compare the proposed method against several relevant state-of-the-art methods using four benchmark databases containing user generated videos (CVD2014, KoNViD-1k, LIVE-Qualcomm, and LIVE-VQC). The results show that the proposed method at a substantially lower computational cost predicts subjective video quality in line with the state of the art methods on individual databases and generalizes better than existing methods in cross-database setup.
In this paper we analyze the most used measures for the assessment of spectral similarity of reflectance and radiance signals. First of all we divide them in five groups on the basis of the type of errors they measure. We proceed analyzing their mathematical definition to identify unintended behaviors and types of errors they are blind to. Then exploiting the Munsell atlas we analyze the correlation between metrics in terms of both Pearson's Linear Correlation Coefficient (PLCC) and Spearman's Rank Order Correlation Coefficient (SROCC). Finally we analyze the behaviour of the selected metrics with respect to two different color properties: the Chroma and the Lightness computed in the CIE L* a* b* color space.The source code of the spectral measures considered is available at the following link: https://celuigi.github.io/spectral-similarity-metrics-comparison/.
Large-scale self-supervised pretraining of deep learning models is known to be critical in several fields, such as language processing, where its has led to significant breakthroughs. Indeed, it is often more impactful than architectural designs. However, the use of self-supervised pretraining lags behind in several domains, such as hyperspectral images, due to data scarcity. This paper addresses the challenge of data scarcity in the development of methods for spatial super-resolution of hyperspectral images (HSI-SR). We show that state-of-the-art HSI-SR methods are severely bottlenecked by the small paired datasets that are publicly available, also leading to unreliable assessment of the architectural merits of the models. We propose to capitalize on the abundance of high resolution (HR) RGB images to develop a self-supervised pretraining approach that significantly improves the quality of HSI-SR models. In particular, we leverage advances in spectral reconstruction methods to create a vast dataset with high spatial resolution and plausible spectra from RGB images, to be used for pretraining HSI-SR methods. Experimental results, conducted across multiple datasets, report large gains for state-of-the-art HSI-SR methods when pretrained according to the proposed procedure, and also highlight the unreliability of ranking methods when training on small datasets.