A frame-based ANN for classification of hyperspectral images: assessment of mechanical damage in mushrooms

Machine learning-enabled hyperspectral approaches for structural characterization of precooked noodles during refrigerated storage

The structural features of precooked noodles during refrigerated storage were non-destructively characterized using hyperspectral imaging (HSI) technology along with conventional analytical methods. The precooked noodles displayed a more rigid texture and restricted water mobility over the storage period, derived from the recrystallization of starch. Dimensionality reduction techniques revealed robust correlations between the storage duration and HSI absorbance of the noodles, and from their loading plots, the specific peaks of the noodles related to their structural changes were identified at wavelengths of around 1160 and 1400 nm. The strong relationships between the HSI results of the noodles and their storage period/texture were confirmed by training four machine learning models on the HSI data. In particular, the support vector algorithm displayed the best prediction performance for classifying precooked noodles by storage period (98.3% accuracy) and for predicting the noodle texture (R2 = 0.914).

Application of Identification and Evaluation Techniques for Edible Mushrooms: A Review

Edible mushrooms are healthy food with high nutritional value, which is popular with consumers. With the increase of the problem of mushrooms being confused with the real and pollution in the market, people pay more and more attention to food safety. More than 167 articles of edible mushroom published in the past 20 years were reviewed in this paper. The analysis tools and data analysis methods of identification and quality evaluation of edible mushroom species, origin, mineral elements were reviewed. Five techniques for identification and evaluation of edible mushrooms were introduced and summarized. The macroscopic, microscopic and molecular identification techniques can be used to identify species. Chromatography, spectroscopy technology combined with chemometrics can be used for qualitative and quantitative study of mushroom and evaluation of mushroom quality. In addition, multiple supervised pattern-recognition techniques have good classification ability. Deep learning is more and more widely used in edible mushroom, which shows its advantages in image recognition and prediction. These techniques and analytical methods can provide strong support and guarantee for the identification and evaluation of mushroom, which is of great significance to the development and utilization of edible mushroom.

Impact of post-harvest processing or thermal dehydration on physiochemical, nutritional and sensory quality of shiitake mushrooms

Shiitake mushrooms are one of the most popular and highly consumed mushrooms worldwide both in fresh and dry forms. However, it rapidly starts losing its quality immediately after harvest which necessitates processing and/or proper storage before being distributed. However, the processes used for preserving other mushrooms (e.g., Agaricus) become unviable for shiitake due to its uniqueness (higher respiration rate, varied biochemicals, growth, etc.) which demands individual studies on shiitake. This review starts by listing the factors and their interdependence leading to a quality decline in shiitake after harvest. Understanding well about these factors, numerous post-harvest operations preserve shiitake as fresh form for a shorter period and as dried forms for a longer shelf-life. These processes also affect the intrinsic quality and nutrients of shiitake. This review comprehensively summarizes and discusses the effects of chemical processing (washing, fumigation, coating, and ozone), modified atmosphere packaging (including irradiation) on the quality of fresh shiitake while discussing their efficiency in extending their shelf-life by inhibiting microbial spoilage and deterioration in quality including texture, appearance, nutrients, and favor. It also reviews the impact of thermal dehydration on the quality of dried shiitake mushrooms, especially the acquired unique textural, nutritional, and aromatic properties along with their merits and limitations. Since shiitake are preferred to be low-cost consumer products, the applicability of freeze-drying and sophisticated novel methodologies, which prove to be expensive and/or complex, are discussed. The review also outlines the challenges and proposes the subsequent future directives, which either retains/enhances the desirable quality in shiitake mushrooms.

Machine learning techniques for analysis of hyperspectral images to determine quality of food products: A review

Non-destructive testing techniques have gained importance in monitoring food quality over the years. Hyperspectral imaging is one of the important non-destructive quality testing techniques which provides both spatial and spectral information. Advancement in machine learning techniques for rapid analysis with higher classification accuracy have improved the potential of using this technique for food applications. This paper provides an overview of the application of different machine learning techniques in analysis of hyperspectral images for determination of food quality. It covers the principle underlying hyperspectral imaging, the advantages, and the limitations of each machine learning technique. The machine learning techniques exhibited rapid analysis of hyperspectral images of food products with high accuracy thereby enabling robust classification or regression models. The selection of effective wavelengths from the hyperspectral data is of paramount importance since it greatly reduces the computational load and time which enhances the scope for real time applications. Due to the feature learning nature of deep learning, it is one of the most promising and powerful techniques for real time applications. However, the field of deep learning is relatively new and need further research for its full utilization. Similarly, lifelong machine learning paves the way for real time HSI applications but needs further research to incorporate the seasonal variations in food quality. Further, the research gaps in machine learning techniques for hyperspectral image analysis, and the prospects are discussed.

•

Artificial neural network has been intensively used for Hyperspectral image (HSI) analysis.

•

Support vector machines and random forest techniques are gaining momentum for HSI analysis.

•

Deep learning applications has potential for implementation in real time HSI analysis.

•

Lifelong machine learning needs further research to incorporate the seasonal variations in food quality.

Hyperspectral Image Classification Using Parallel Autoencoding Diabolo Networks on Multi-Core and Many-Core Architectures

One of the most important tasks in hyperspectral imaging is the classification of the pixels in the scene in order to produce thematic maps. This problem can be typically solved through machine learning techniques. In particular, deep learning algorithms have emerged in recent years as a suitable methodology to classify hyperspectral data. Moreover, the high dimensionality of hyperspectral data, together with the increasing availability of unlabeled samples, makes deep learning an appealing approach to process and interpret those data. However, the limited number of labeled samples often complicates the exploitation of supervised techniques. Indeed, in order to guarantee a suitable precision, a large number of labeled samples is normally required. This hurdle can be overcome by resorting to unsupervised classification algorithms. In particular, autoencoders can be used to analyze a hyperspectral image using only unlabeled data. However, the high data dimensionality leads to prohibitive training times. In this regard, it is important to realize that the operations involved in autoencoders training are intrinsically parallel. Therefore, in this paper we present an approach that exploits multi-core and many-core devices in order to achieve efficient autoencoders training in hyperspectral imaging applications. Specifically, in this paper, we present new OpenMP and CUDA frameworks for autoencoder training. The obtained results show that the CUDA framework provides a speed-up of about two orders of magnitudes as compared to an optimized serial processing chain.

Joint Alternate Small Convolution and Feature Reuse for Hyperspectral Image Classification

A hyperspectral image (HSI) contains fine and rich spectral information and spatial information of ground objects, which has great potential in applications. It is also widely used in precision agriculture, marine monitoring, military reconnaissance and many other fields. In recent years, a convolutional neural network (CNN) has been successfully used in HSI classification and has provided it with outstanding capacity for improving classification effects. To get rid of the bondage of strong correlation among bands for HSI classification, an effective CNN architecture is proposed for HSI classification in this work. The proposed CNN architecture has several distinct advantages. First, each 1D spectral vector that corresponds to a pixel in an HSI is transformed into a 2D spectral feature matrix, thereby emphasizing the difference among samples. In addition, this architecture can not only weaken the influence of strong correlation among bands on classification, but can also fully utilize the spectral information of hyperspectral data. Furthermore, a 1 × 1 convolutional layer is adopted to better deal with HSI information. All the convolutional layers in the proposed CNN architecture are composed of small convolutional kernels. Moreover, cascaded composite layers of the architecture consist of 1 × 1 and 3 × 3 convolutional layers. The inputs and outputs of each composite layer are stitched as the inputs of the next composite layer, thereby accomplishing feature reuse. This special module with joint alternate small convolution and feature reuse can extract high-level features from hyperspectral data meticulously and comprehensively solve the overfitting problem to an extent, in order to obtain a considerable classification effect. Finally, global average pooling is used to replace the traditional fully connected layer to reduce the model parameters and extract high-dimensional features from the hyperspectral data at the end of the architecture. Experimental results on three benchmark HSI datasets show the high classification accuracy and effectiveness of the proposed method.