Food recognition using neural network classifier and multiple hypotheses image segmentation

ABSTRACT

This paper proposes the multiple-hypotheses image segmentation and feed-forward neural network classifier for food recognition to improve the performance. Initially, the food or meal image is given as input. Then, the segmentation is applied to identify the regions, where a particular food item is located using salient region detection, multi-scale segmentation, and fast rejection. Then, the features of every food item are extracted by the global feature and local feature extraction. After the features are obtained, the classification is performed for each segmented region using a feed-forward neural network model. Finally, the calorie value is computed with the aid of (i) food volume and (ii) calorie and nutrition measure based on mass value. The experimental results and performance evaluation are validated. The outcome of the proposed method attains 0.947 for Macro Average Accuracy (MAA) and 0.959 for Standard Accuracy (SA), which provides better classification performance.     

Micron-Scale Soft Actuators Fabricated from Multi-Shell Polystyrene Particle-Gold Nanoparticle Nanohybrids

Actuators made of soft matter are needed for a variety of fields ranging from biomedical devices to soft robotics to microelectromechanical systems. While there are a variety of excellent methods of soft actuation known, the field is still an area of intense research activity as new niches and needs emerge with new technology development. Here, a soft actuation system is described, based on a core-multi-shell particle, which moves via photothermal expansion. The system consists of a novel polystyrene-based thermally expandable microsphere, with a secondary shell of a silicate-silane graft copolymer, to which gold nanoparticles are covalently linked. The gold nanoparticles act as photothermal nano-transducers, converting light energy into the thermal energy necessary for microsphere expansion, which in turn results in material movement. Actuation is shown in isolated particles in thermal and photothermal regimes using metal ceramic heaters or 520 nm laser illumination, respectively. Macroscale actuation is demonstrated by making a composite material of particles suspended in the transparent elastomer polydimethylsiloxane. The sample demonstrates an inchworm-like movement by starting from an arched geometry. Overall, this work describes a new particle-based actuation method for soft materials, and demonstrates its utility in driving the movement of a composite elastomer.     

Mammogram Learning System for Breast Cancer Diagnosis Using Deep Learning SVM

The most common form of cancer for women is breast cancer. Recent advances in medical imaging technologies increase the use of digital mammograms to diagnose breast cancer. Thus, an automated computerized system with high accuracy is needed. In this study, an efficient Deep Learning Architecture (DLA) with a Support Vector Machine (SVM) is designed for breast cancer diagnosis. It combines the ideas from DLA with SVM. The state-of-the-art Visual Geometric Group (VGG) architecture with 16 layers is employed in this study as it uses the small size of 3 × 3 convolution filters that reduces system complexity. The softmax layer in VGG assumes that the training samples belong to exactly only one class, which is not valid in a real situation, such as in medical image diagnosis. To overcome this situation, SVM is employed instead of the softmax layer in VGG. Data augmentation is also employed as DLA usually requires a large number of samples. VGG model with different SVM kernels is built to classify the mammograms. Results show that the VGG-SVM model has good potential for the classification of Mammographic Image Analysis Society (MIAS) database images with an accuracy of 98.67%, sensitivity of 99.32%, and specificity of 98.34%.     

Neural network classifier and multiple hypothesis image segmentation for dietary assessment using calorie calculator

Nowadays, dietary assessment becomes the emerging system for evaluating the person’s food intake. In this paper, the multiple hypothesis image segmentation and feed-forward neural network classifier are proposed for dietary assessment to enhance the performance. Initially, the segmentation is applied to input image which is used to determine the regions where a particular food item is located using salient region detection, multi-scale segmentation, and fast rejection. Then, the significant feature of food items is extracted by the global feature and local feature extraction method. After the features are obtained, the classification is performed for each segmented region using feed-forward neural network model. Finally, the calorie value is computed with the aid of (i) food area volume and (ii) calorie and nutrition measure based on mass value. The outcome of the proposed method attains 96% of accuracy value which provides the better classification performance.     

Multi-scale discriminant saliency with wavelet-based Hidden Markov Tree modelling

Supposed saliency is a binary classification between centre and surround classes, saliency value is measured as their discriminant power. As the features are defined by sizes of chosen windows, a saliency value at each location is varied accordingly. This paper proposes computing saliency as discriminant power in multiple dyadic scales of Wavelet Hidden Markov Tree (HMT), in which two consecutive dyadic scales provide surrounding and central features, organized in a quad-tree structure. Their discriminant power is estimated as maximum a posterior probability (MAP) by Expectation-Maximization (EM) iterations. Then, a final saliency value is the maximum discriminant power generated among these scales. Standard quantitative tools and qualitative assessments are used for evaluating the proposed multi-scale discriminant saliency (MDIS) against the well-know information based approach AIM on its image collection with eye-tracking data. Simulation results are presented and analysed to verify the validity of MDIS as well as point out its limitation for further research direction.     

Does power posing affect gerontechnology adoption among older adults?

Recognising the rise of an aging population and independent living among older adults, many governments and organisations have developed and promoted new technologies in the form of gerontechnologies to support the needs and enhance the well-being of older adults. However, the adoption of products using such technology remains modest among the aging population. This study introduces the notion of power in the form of power posing and examines its impact on new technology adoption, particularly gerontechnology, among older adults. Using an experimental approach on a sample of older adults exposed to an in-house near-field communication-enabled light system, the study finds that high-power poses have a greater and more positive impact on older adults’ perceived ease of use of, perceived usefulness of, and intentions to use gerontechnology than low-power poses. Implications, limitations, and future research directions are discussed.     

Poly(3-hexylthiophene) and hexafluoro-2-propanol-substituted polysiloxane based OFETs as a sensor for explosive vapor detection

The organic field effect transistors (OFETs) with regioregular poly 3-hexylthiophene (rr-P3HT) and hexafluoro-2-propanol-substituted polysiloxane (SXFA) as an organic layer, have been used for detection of explosive vapors with excellent sensitivity of less than 70 ppt for 1,3,5-trinitro-1,3,5-triazacyclohexane (RDX) and less than 100 ppt for 2,4,6-trinitrotoluene (TNT). The sensor response (% change in saturation current) was found to be 125 ± 10% for TNT and 90 ± 10% for RDX. It was also observed that the incorporation of Cu^II tetraphenylporphyrin (CuTPP) into rr-P3HT/SXFA matrix resulted in an improved selectivity for the vapors of nitro based analytes (TNT, RDX and DNB) as compared to the vapors of non explosive oxidizing agents such as nitrobenzene (NB), benzoquinone (BQ) and benzophenone (BP). This is attributed to the increased binding of the vapors containing nitro compound to the thin films due to the presence of CuTTP. Spin coated thin films were further characterized by Atomic Force Microscopy (AFM) and Electrostatic Force Microscopy (EFM).     

Action-research-based optimisation model for health care behaviour change in rural India

In this research, we utilise a multi-method braided approach for a real-world health care intervention in rural India. We first use action research to conceptually underpin the role of self-help groups, disease profiles and appropriate interventions in health care behaviour change and improvement. Action research is then coupled with a non-linear integer-programming-based simulation to determine how health care resources need to be distributed over time, disease type and groups. The action research provides specific research questions and distributional forms needed for the optimisation analysis, while the non-linear integer-programming simulation provides specific recommendations on where and when to distribute resources. External validity of the study is ensured by the adoption of recommendations in the field as part of the action research. Findings from this research have important policy implications. For example, it identifies the value of continuous embedded interventions as opposed to periodic ones and the importance of flexibility in terms of which community groups and disease types to serve over time. This research also provides an important bridge between the operations research, health care in developing nations, organisational theory and humanitarian operations bodies of knowledge.