Separation of closely related structural isomers by inclusion complexation with tailor-made host compounds. Crystal structures of 2,2-di (p-hydroxyphenyl) propane and Its 1:1 complex withp-cresol

The crystal structures of the 2,2-di(p-hydroxyphenyl)propane host and its 1:1 adducts withm-andp-cresol guests have been studied. The preferential complexation of this host withp-cresol overm-cresol is related to the opposite trend exhibited by 1,1-di(p-hydroxyphenyl)cyclohexane; both hosts can separate effectively the two cresols from their liquid mixture by crystalline inclusion. A plausible explanation of the different inclusion features is provided by examining the intermolecular association in the corresponding solids. The analysed structures are stabilized by strong and continuous H-bonding between the constituent entities along two dimensions, and by weak van der Waals forces along the third axis. The p-cresol complex of the title host reveals a unique arrangement within and a more efficient packing of the layered structure, and thus represents a more stable and less soluble crystal lattice than itsm-cresol analog. Supplementary Data relating to this article are deposited with the British Library as Supplementary Publication No. SUP 82099 (8 pages).     

Multipoint recognition of catecholamines by hydrindacene-based receptors accompanied by the complexation-induced conformational switching

The molecular recognition of catecholamines by hydrindacene-based receptors 1 and 2, as well as the durene-based receptor 3, and the guest-induced conformational changes are reported. These receptors selectively bind adrenaline and dopamine salts through the guests' ammonium group and 3-hydroxyl group on the aromatic ring. In the case of adrenaline, an additional hydrogen bond with a benzylic hydroxyl group is formed. In 2 % CD3CN/CDCl3, the association constants are of the order of 10(4) M(-1), which is much larger than with guests without the 3-hydroxyl groups (10(3) M(-1)). The two amide groups of receptor 1 can rotate freely around the C(aromatic)--C(amide) bond, whereas the tert-amide in 2 changes between two stable conformations at a slow enough rate to allow detection by (1)H NMR spectroscopy. In the absence of a guest molecule, the syn-conformer is less stable than the anti-conformer. On complex formation with adrenaline, the syn-conformer becomes dominant due to an intramolecular dipole-reversal effect in addition to multipoint hydrogen bonding.     

Molecule-Responsive Fluorescent Sensors of α-Helix Peptides Bearing α-Cyclodextrin, Pyrene and Nitrobenzene Units in Their Side Chains

-Helix peptides bearing one unit of -cyclodextrin (-CD), one unit of pyrene and one unit of nitrobenzene (NB) in their side chains have been designed and synthesized as novel molecule-responsive devices.In both the CD-peptides, -PR17 and -PL17, the NB unit is separated from the CD unit by two turns of the helix. Two reference peptides (PL17, and -P17,) have also been synthesized. The circular dichroism studies in the peptide absorption region (200–250 nm)of -PR17 and -PL17 suggestthat the CD-peptides form stable-helixstructures (83–77%), which was destabilized by accommodating guest molecules (e.g., n-pentanol) into the CD cavity. It suggests that formation of intramolecular host–guest(CD–NB) complex stabilized thehelical structure and exogenous guest molecule excluded the appending NB moiety from inside to outside of the CD cavity, thereby causing destabilization of the helical structure and increasing the random coil content. The ICD spectra of the peptides in the pyrene and nitrobenzene absorption region (250–40 nm) suggest that NB forms inclusion complex with CD. The fluorescence studies revealed that the fluorescence of the pyrene unit is quenched by the NB unit in -PR17 and -PL17. The fluorescence intensity increases with increasing guest concentration for the CD-peptides.This guest-responsiveenhancement in the fluorescence intensity can be explained in terms of increased distance between the pyrene and NB moieties, which is caused by exclusion of the NB moiety from the CD cavity by guest accommodation. Using the guest-responsive fluorescence quenching properties of the CD-peptides, we have obtained binding constants for various short chain alkanols. -PL17 has higher binding affinity to the guest molecules than its isomer, -PR17, indicating that the location of functional groups on the peptide scaffold is important in molecule detection.     

Preparation of Saccharide-Silica Composites by a Sol-Gel Method and the Application for Optical Resolution of Metal Chelate Compounds

A novel organic-inorganic composite was prepared by a sol-gel process using hydrolysis and condensation of tetraethoxysilane (TEOS) reacted with a saccharide. During the process, ethoxy groups of the TEOS were replaced with hydroxyl groups of the saccharide and, consequently, saccharide molecules were combined with silica through their hydroxyl groups. Samples obtained under different reaction conditions were used for optical resolution of a metal chelate compound. The resolution ability of the composites was affected by the amount of water added for hydrolysis and of the saccharide. In particular, the amount of water drastically influenced the optical resolution performance. Composites obtained at the lower H2O/TEOS ratio gave the metal chelate compound a higher optical rotation. Three types of composites with the same composition were prepared by kneading, impregnation and by the sol-gel method, and were compared. The sol-gel composite showed the best optical resolution ability. It was concluded that the optical resolution ability was closely related to dispersibility of saccharide in the silica matrix.     

“模式识别”课程线上线下混合教学模式探索与实践

在模式识别课程教学中,针对传统线下课程短板,开展线上与线下相结合的混合教学模式教学改革探索。阐述了本课程教学改革的教学环节设计、教学实施方法、考核方式等,以思维导图和综合案例串联庞杂知识点,采用多样化手段,加强对学生创新思维和能力的培养。教学实践证明,该课程教学改革调动了学生主动性,培育了其创新能力与科学精神,丰富了线上线下混合教学改革的理论研究。     

Deep ensemble-based hard sample mining for food recognition

Deep neural networks represent a compelling technique to tackle complex real-world problems, but are over-parameterized and often suffer from over- or under-confident estimates. Deep ensembles have shown better parameter estimations and often provide reliable uncertainty estimates that contribute to the robustness of the results. In this work, we propose a new metric to identify samples that are hard to classify. Our metric is defined as coincidence score for deep ensembles which measures the agreement of its individual models. The main hypothesis we rely on is that deep learning algorithms learn the low-loss samples better compared to large-loss samples. In order to compensate for this, we use controlled over-sampling on the identified ”hard” samples using proper data augmentation schemes to enable the models to learn those samples better. We validate the proposed metric using two public food datasets on different backbone architectures and show the improvements compared to the conventional deep neural network training using different performance metrics.     

Joint frame rate adaptation and object recognition model selection for stabilized unmanned aerial vehicle surveillance

We propose an adaptive unmanned aerial vehicle (UAV)-assisted object recognition algorithm for urban surveillance scenarios. For UAV-assisted surveillance, UAVs are equipped with learning-based object recognition models and can collect surveillance image data. However, owing to the limitations of UAVs regarding power and computational resources, adaptive control must be performed accordingly. Therefore, we introduce a self-adaptive control strategy to maximize the time-averaged recognition performance subject to stability through a formulation based on Lyapunov optimization. Results from performance evaluations on real-world data demonstrate that the proposed algorithm achieves the desired performance improvements.     

Extracting embedded messages using adaptive steganography based on optimal syndrome-trellis decoding paths

Privacy protection is the key to maintaining the Internet of Things (IoT) communication strategy. Steganography is an important way to achieve covert communication that protects user data privacy. Steganalysis technology is the key to checking steganography security, and its ultimate goal is to extract embedded messages. Existing methods cannot extract under known cover images. To this end, this paper proposes a method of extracting embedded messages under known cover images. First, the syndrome-trellis encoding process is analyzed. Second, a decoding path in the syndrome trellis is obtained by using the stego sequence and a certain parity-check matrix, while the embedding process is simulated using the cover sequence and parity-check matrix. Since the decoding path obtained by the stego sequence and the correct parity-check matrix is optimal and has the least distortion, comparing the path consistency can quickly filter the coding parameters to determine the correct matrices, and embedded messages can be extracted correctly. The proposed method does not need to embed all possible messages for the second time, improving coding parameter recognition significantly. The experimental results show that the proposed method can identify syndrome-trellis coding parameters in stego images embedded by adaptive steganography quickly to realize embedded message extraction.     

增量式块主成分分析的焊缝图像特征提取算法北大核心CSCD

针对焊缝图像特征提取的实时性问题,该文提出一种增量式块主成分分析(incremental block principal component analysis,IBlockPCA)算法,用于焊缝特征主成分的提取。该算法先将焊缝表面图像分割成子图像块并对其进行重构,然后利用提出的IBlockPCA算法对局部块图像进行增量式特征提取,并采用KNN算法对提取的特征主成分进行分类识别;最后在焊缝数据集上进行了算法的性能对比。实验结果表明,该算法在收敛率、分类率及复杂度等方面均优于其他主成分分析(principal component analysis,PCA)算法,其分类识别率为97.5%,其平均处理速度可达50 frame/s,能够满足焊缝表面图像的实时性处理需求。     

基于LRC与多样本扩充的指静脉识别方法北大核心CSCD

针对目前手指静脉识别由于训练样本不足引起图像识别率低的问题,提出基于线性回归分类(linear regression classification,LRC)与多样本扩充的指静脉识别方法。首先,利用矩阵变换生成原始图像的镜像,训练原始图像与镜像,增加指静脉图像中包含的有用信息;然后,基于LRC对测试和训练样本进行分类;最后,通过计算偏差得到最终分类结果,求出识别率。此外,设计了一种指静脉采集装置收集得到自建指静脉数据库。实验结果表明:所提算法在自建指静脉数据库、山东大学指静脉数据库、马来西亚理工大学指静脉数据库上的识别率分别达到98.93%、98.89%、99.67%,最低等误率为2.3888%。实验结果与其他传统和流行算法相比具有明显优势,拥有良好的实际应用价值。