Automatic fruit detection is a very important benefit of harvesting robots. However, complicated environment conditions, such as illumination variation, branch, and leaf occlusion as well as tomato overlap, have made fruit detection very challenging. In this study, an improved tomato detection model called YOLO-Tomato is proposed for dealing with these problems, based on YOLOv3. A dense architecture is incorporated into YOLOv3 to facilitate the reuse of features and help to learn a more compact and accurate model. Moreover, the model replaces the traditional rectangular bounding box (R-Bbox) with a circular bounding box (C-Bbox) for tomato localization. The new bounding boxes can then match the tomatoes more precisely, and thus improve the Intersection-over-Union (IoU) calculation for the Non-Maximum Suppression (NMS). They also reduce prediction coordinates. An ablation study demonstrated the efficacy of these modifications. The YOLO-Tomato was compared to several state-of-the-art detection methods and it had the best detection performance.

An electrical impedance spectroscopy (EIS) experiment was performed using a double-shell electrical model to investigate the feasibility of detecting physiological changes in lettuce leaves over 16 h. Four lettuce plants were used, and the impedance spectra of the leaves were measured five times per plant every hour at frequencies of 500 Hz and 300 kHz. Estimated R-C parameters were computed, and the results show that the lettuce leaves closely fit the double-shell model (DSM). The average resistance ratios of R₁ = 10.66R₄ and R₁ = 3.34R₂ show high resistance in the extracellular fluid (ECF). A rapid increase in resistance (R₁, R₂, and R₄) and a decrease in capacitance (C₃ and C₅) during water uptake were observed. In contrast, a gradual decrease in resistance and an increase in capacitance were observed while the LED light was on. Comparative studies of leaf physiology and electrical value changes support the idea that EIS is a great technique for the early monitoring of plant growth for crop production.

When it comes to growing lettuce, specific nutrients play vital roles in its growth and development. These essential nutrients include full nutrients (FN), nitrogen (N), phosphorus (P), and potassium (K). Insufficient or excess levels of these nutrients can have negative effects on lettuce plants, resulting in various deficiencies that can be observed in the leaves. To better understand and identify these deficiencies, a deep learning approach is employed to improve these tasks. For this study, YOLOv8 Nano, a lightweight deep network, is chosen to classify the observed deficiencies in lettuce leaves. Several enhancements to the baseline algorithm are made, the backbone is replaced with VGG16 to improve the classification accuracy, and depthwise convolution is incorporated into it to enrich the features while keeping the head unchanged. The proposed network, incorporating these modifications, achieved superior classification results with a top-1 accuracy of 99%. This method outperformed other state-of-the-art classification methods, demonstrating the effectiveness of the approach in identifying lettuce deficiencies. The objective of this research was to improve the baseline algorithm to complete the classification task with a top-1 accuracy above 85%, a FLOP inferior to 10G, and classification latency below 170 ms per image.

Plant and fruit diseases significantly impact agricultural economies by diminishing crop quality and yield. Developing precise, automated detection techniques is crucial to minimize losses and drive economic growth. We introduce YOLO-AppleScab, integrating Content-Aware ReAssembly of FEature ($\mathbb{C}\mathbb{A}\mathbb{R}\mathbb{A}\mathbb{F}\mathbb{E}$) architecture into YOLOv7 for enhanced apple fruit detection and disease classification. The model achieves impressive metrics: F1, recall, and precision of 89.75%, 85.20%, and 94.80%, and a mean average precision of 89.30% at $IoU$ 0.5. With 64% efficiency, this model’s integration with YOLOv7 improves detection, promising economic benefits by accurately detecting apple scab disease and reducing agricultural damage.

In this paper, we demonstrated a polarimetric fiber vibration sensor based on a polarization-diversified loop (PDL) with short polarization-maintaining photonic crystal fiber (PM-PCF) as a sensor head and investigated the dependence of its frequency response on the sensing fiber length. The fabricated PM-PCF vibration sensor is composed of a polarization beam splitter for forming the PDL, PM-PCF as a sensor head, and two wave plates such as a quarter-wave plate and a half-wave plate. A laser diode and a photodetector were also utilized for intensity-based vibration measurement. In particular, two kinds of PM-PCF segments with fiber lengths of ~6 and ~10 cm were employed to explore the effect of the sensor head length on the sensor frequency response. The frequency response was examined over 1–3000 Hz using a piezoelectric transducer that can apply single-frequency vibration to the sensor head. Every measured frequency response showed a resonance peak at each sensor head length, and cut-off frequencies were measured as ~1240 and ~794 Hz for PM-PCF lengths of ~6 and ~10 cm, respectively. That is, a higher cut-off frequency was obtained in a shorter sensor head length. Moreover, it was observed that the sensor head length dependence of the frequency response was maximized in a frequency range from 1000 to 2400 Hz and faded away beyond 2400 Hz. From the measurement results, it is concluded that the short sensor head can provide higher normalized sensitivity, i.e., better signal to noise ratio, at frequencies higher than the resonance frequency, let alone convenience of installation and in sensitiveness to external perturbations in a PDL-based PM-PCF vibration sensor. The phase shift per unit strain and the minimum detectable strain perturbation, regarded as sensitivity and resolution, were measured as ~0.377 mrad/με and ~0.16 nε/Hz^1/2 at 2000 Hz for ~6-cm-long PM-PCF, respectively.

A second-order fiber interleaving filter that can provide a channel interleaving capability in second-order comb spectra is proposed and demonstrated on the basis of a polarization-diversified loop comprised of a four-port polarization beam splitter, four rotatable half-wave plates (HWPs), and three high birefringence fiber (HBF) segments. Each HBF segment is positioned between two adjoining HWPs, and two HWPs located among the three HBF segments determine the effective angular orientation difference among their principal axes, or fast and slow axes. At specific orientation angles of the four HWPs, second-order comb spectra such as flat-top and narrow-band ones could be obtained with a free spectral range (FSR) of ∼ 0.792 nm . In particular, the frequency interleaving of these second-order comb spectra, that is, a half FSR switching, could be implemented by controlling two out of four HWPs, which is the first demonstration of the interleaving operation in second-order comb spectra without expensive birefringence modulators. These spectral characteristics of the proposed filter were theoretically predicted and experimentally demonstrated.