With the rapid development of the Internet, watermarks are widely used in images to protect copyright. This implies that the robustness of watermark is very important. In recent years, there have been some studies to evaluate watermark performance by removing the watermark. Among them, some methods need to mark the watermark position in advance, and some require multiple images with the same watermark. Moreover, when the colour of thewatermark is similar to that of the background, the existing methods can hardly remove the watermark from the watermarked image. In the proposed work, the authors presented a watermark removal structure consisting of watermark extraction and image inpainting to address the aforementioned issues. In particular, the extraction network is used to extract the watermark in the watermarked image, and the inpainting network is used to inpainting image for a better watermark removal image, respectively. Finally, the authors train and test the developed network architecture by constructing two data sets, i.e. white watermarked image data set (WW-data set) and colour watermarked image data set (CW-data set). The proposed method not only has better performance on the WW-data set than the current latest methods (on the CW-data set, other methods have almost failed) but also effectively removes the watermarks.

Time–temperature indicators (TTIs) can monitor the quality and safety of food. A new temperature–time point comparison method was proposed to match TTIs with food. This method omits the step of calculating activation energy (Ea). It only compares the difference between TTI response time and food shelf life to determine their matching degree. Taking gold nanoparticle‐based TTIs and muffins as experimental objects, the new and the traditional matching methods were used to match the absorbance of TTI and the peroxide value of muffins. The two results are not significantly different. TTIs with gelatin solution and gold precursor solution concentration of 150.00  and 2.05 mg/mL, respectively, can show the quality of muffins. TTIs changed from light yellow to pink and finally appeared deep purple. The deep purple represented spoilage and inedibility of muffins. Comparing Ea of food and that of TTIs can preliminarily evaluate their matching degree, improving the experiment efficiency. Hence, it is reasonable to use the traditional matching method in most cases, and use the new method only when Ea of food cannot be obtained. Practical Application The deterioration rate of food is usually calculated by developing kinetic models of characteristic quality parameters. When the reaction rate is unavailable or inaccurate, the activation energy of food cannot be obtained. In this case, it is impossible to match TTIs with food based on the traditional method. This research develops a new matching method and helps TTIs and food to be matched without considering activation energy. It will promote the application of TTIs in more products.

Methamphetamine is a highly addictive drug of abuse, which will cause a series of abnormal consequences mentally and physically. This paper is aimed at studying whether the abnormalities of regional homogeneity (ReHo) could be effective features to distinguish individuals with methamphetamine dependence (MAD) from control subjects using machine-learning methods. We made use of resting-state fMRI to measure the regional homogeneity of 41 individuals with MAD and 42 age- and sex-matched control subjects and found that compared with control subjects, individuals with MAD have lower ReHo values in the right medial superior frontal gyrus but higher ReHo values in the right temporal inferior fusiform. In addition, AdaBoost classifier, a pretty effective ensemble learning of machine learning, was employed to classify individuals with MAD from control subjects with abnormal ReHo values. By utilizing the leave-one-out cross-validation method, we got the accuracy more than 84.3%, which means we can almost distinguish individuals with MAD from the control subjects in ReHo values via machine-learning approaches. In a word, our research results suggested that the AdaBoost classifier-neuroimaging approach may be a promising way to find whether a person has been addicted to methamphetamine, and also, this paper shows that resting-state fMRI should be considered as a biomarker, a noninvasive and effective assistant tool for evaluating MAD.

Few studies have investigated the functional patterns of methamphetamine abstainers. A better understanding of the underlying neurobiological mechanism in the brains of methamphetamine abstainers will help to explain their abnormal behaviors. Forty-two male methamphetamine abstainers, currently in a long-term abstinence status (for at least 14 months), and 32 male healthy controls were recruited. All subjects underwent functional MRI while responding to drug-associated cues. This study proposes to combine a convolutional neural network with a short-time Fourier transform to identify different brain patterns between methamphetamine abstainers and controls. The short-time Fourier transformation provides time-localized frequency information, while the convolutional neural network extracts the structural features of the time–frequency spectrograms. The results showed that the classifier achieved a satisfactory performance (98.9% accuracy) and could extract robust brain voxel information. The highly discriminative power voxels were mainly concentrated in the left inferior orbital frontal gyrus, the bilateral postcentral gyri, and the bilateral paracentral lobules. This study provides a novel insight into the different functional patterns between methamphetamine abstainers and healthy controls. It also elucidates the pathological mechanism of methamphetamine abstainers from the view of time–frequency spectrograms.

No abstract