Improvement in Purity of Healthy Tomato Seeds Using an Image-Based One-Class Classification Method

A method for detecting the quality of cotton seeds based on an improved ResNet50 model

The accurate and rapid detection of cotton seed quality is crucial for safeguarding cotton cultivation. To increase the accuracy and efficiency of cotton seed detection, a deep learning model, which was called the improved ResNet50 (Impro-ResNet50), was used to detect cotton seed quality. First, the convolutional block attention module (CBAM) was embedded into the ResNet50 model to allow the model to learn both the vital channel information and spatial location information of the image, thereby enhancing the model's feature extraction capability and robustness. The model's fully connected layer was then modified to accommodate the cotton seed quality detection task. An improved LRelu-Softplus activation function was implemented to facilitate the rapid and straightforward quantification of the model training procedure. Transfer learning and the Adam optimization algorithm were used to train the model to reduce the number of parameters and accelerate the model's convergence. Finally, 4419 images of cotton seeds were collected for training models under controlled conditions. Experimental results demonstrated that the Impro-ResNet50 model could achieve an average detection accuracy of 97.23% and process a single image in 0.11s. Compared with Squeeze-and-Excitation Networks (SE) and Coordination Attention (CA), the model's feature extraction capability was superior. At the same time, compared with classical models such as AlexNet, VGG16, GoogLeNet, EfficientNet, and ResNet18, this model had superior detection accuracy and complexity balances. The results indicate that the Impro-ResNet50 model has a high detection accuracy and a short recognition time, which meet the requirements for accurate and rapid detection of cotton seed quality.

Varietal Classification of Lactuca Sativa Seeds Using an Adaptive Neuro-Fuzzy Inference System Based on Morphological Phenes

Seed varieties are often differentiated via the manual and subjective classification of their external textural, spectral, and morphological biosignatures. This traditional method of manually inspecting seeds is inefficient and unreliable for seed phenotyping. The application of computer vision is an ideal solution allied with computational intelligence. This study used Lactuca sativa seed variants, which are commercially known as grand rapid, Chinese loose-leaf, and iceberg (which serves as noise data for extended model evaluation), in determining their corresponding classifications based on the extended morphological phenes using computational intelligence. Red-green-blue (RGB) imaging was employed for individual kernels. Extended morphological phenes, that is, solidity, roundness, compactness, and shape factors, were computed based on seed architectural traits and used as predictors to discriminate among the three cultivars. The suitability of ANFIS, NB, and CT was explored using a limited dataset. A mean accuracy of 100% was manifested in ANFIS; thus, it was proved to be the most reliable model.

Hyperspectral Shortwave Infrared Image Analysis for Detection of Adulterants in Almond Powder with One-Class Classification Method

The widely used techniques for analyzing the quality of powdered food products focus on targeted detection with a low-throughput screening of samples. Owing to potentially significant health threats and large-scale adulterations, food regulatory agencies and industries require rapid and non-destructive analytical techniques for the detection of unexpected compounds present in products. Accordingly, shortwave-infrared hyperspectral imaging (SWIR-HSI) for high throughput authenticity analysis of almond powder was investigated in this study. Two different varieties of almond powder, adulterated with apricot and peanut powder at different concentrations, were imaged using the SWIR-HSI system. A one-class classifier technique, known as data-driven soft independent modeling of class analogy (DD-SIMCA), was used on collected data sets of pure and adulterated samples. A partial least square regression (PLSR) model was further developed to predict adulterant concentrations in almond powder. Classification results from DD-SIMCA yielded 100% sensitivity and 89-100% specificity for different validation sets of adulterated samples. The results obtained from the PLSR analysis yielded a high determination coefficient (R2) and low error values (<1%) for each variety of almond powder adulterated with apricot; however, a relatively higher error rates of 2.5% and 4.4% for the two varieties of almond powder adulterated with peanut powder, which indicates the performance of quantitative analysis model could vary with sample condition, such as variety, originality, etc. PLSR-based concentration mapped images visually characterized the adulterant (apricot) concentration in the almond powder. These results demonstrate that the SWIR-HSI technique combined with the one-class classifier DD-SIMCA can be used effectively for a high-throughput quality screening of almond powder regarding potential adulteration.

Non-Targeted Detection of Adulterants in Almond Powder Using Spectroscopic Techniques Combined with Chemometrics

Methods that combine targeted techniques and chemometrics for analyzing food authenticity can only facilitate the detection of predefined or known adulterants, while unknown adulterants cannot be detected using such methods. Therefore, the non-targeted detection of adulterants in food products is currently in great demand. In this study, FT-IR and FT-NIR spectroscopic techniques were used in combination with non-targeted chemometric approaches, such as one-class partial least squares (OCPLS) and data-driven soft independent modeling of class analogy (DD-SIMCA), to detect adulterants in almond powder adulterated with apricot and peanut powders. The reflectance spectra of 100 pure almond powder samples from two different varieties (50 each) were collected to develop a calibration model based on each spectroscopic technique; each model was then evaluated for four independent sets of two varieties of almond powder samples adulterated with different concentrations of apricot and peanut powders. Classification using both techniques was highly sensitive, the OCPLS approach yielded 90-100% accuracy in different varieties of samples with both spectroscopic techniques, and the DD-SIMCA approach achieved the highest accuracy of 100% when used in combination with FT-IR in all validation sets. Moreover, DD-SIMCA, combined with FT-NIR, achieved a detection accuracy between 91% and 100% for the different validation sets and the misclassified samples belong to the 5% and 7% adulteration sets. These results suggest that spectroscopic techniques, combined with one-class classifiers, can be used effectively in the high-throughput screening of potential adulterants in almond powder.