Sub-Nanogram Resolution Measurement of Inertial Mass and Density Using Magnetic-Field-Guided Bubble Microthruster.

Artificial micro/nanomotors using active particles hold vast potential in applications such as drug delivery and microfabrication. However, upgrading them to micro/nanorobots capable of performing precise tasks with sophisticated functions remains challenging. Bubble microthruster (BMT) is introduced, a variation of the bubble-driven microrobot, which focuses the energy from a collapsing microbubble to create an inertial impact on nearby target microparticles. Utilizing ultra-high-speed imaging, the microparticle mass and density is determined with sub-nanogram resolution based on the relaxation time characterizing the microparticle's transient response. Master curves of the BMT method are shown to be dependent on the viscosity of the solution. The BMT, controlled by a gamepad with magnetic-field guidance, precisely manipulates target microparticles, including bioparticles. Validation involves measuring the polystyrene microparticle mass and hollow glass microsphere density, and assessing the mouse embryo mass densities. The BMT technique presents a promising chip-free, real-time, highly maneuverable strategy that integrates bubble microrobot-based manipulation with precise bioparticle mass and density detection, which can facilitate microscale bioparticle characterizations such as embryo growth monitoring.

Manipulation of Contact Angle Hysteresis at Electrified Ionic Liquid-Solid Interfaces.

Room-temperature ionic liquids (RTILs) are intriguing fluids that have drawn much attention in applications ranging from tribology and catalysis to energy storage. With strong electrostatic interaction between ions, their interfacial behaviors can be modulated by controlling energetics of the electrified interface. In this work, we report atomic-force-microscope measurements of contact angle hysteresis (CAH) of a circular contact line formed on a micron-sized fiber, which is coated with a thin layer of conductive film and intersects an RTIL-air interface. The measured CAH shows a distinct change by increasing the voltage U applied on the fiber surface. Molecular dynamics simulations were performed to illustrate variations of the solidlike layer in the RTIL adsorbed at the electrified interface. The integrated experiments and computations demonstrate a new mechanism to manipulate the CAH by rearrangement of interfacial layers of RTILs induced by the surface energetics.

Research Status in the Use of Surface-Enhanced Raman Scattering (SERS) to Detect Pesticide Residues in Foods and Plant-Derived Chinese Herbal Medicines.

Surface-enhanced Raman scattering (SERS) technology has unique advantages in the rapid detection of pesticides in plant-derived foods, leading to reduced detection limits and increased accuracy. Plant-derived Chinese herbal medicines have similar sources to plant-derived foods; however, due to the rough surfaces and complex compositions of herbal medicines, the detection of pesticide residues in this context continues to rely heavily on traditional methods, which are time consuming and laborious and are unable to meet market demands for portability. The application of flexible nanomaterials and SERS technology in this realm would allow rapid and accurate detection in a portable format. Therefore, in this review, we summarize the underlying principles and characteristics of SERS technology, with particular focus on applications of SERS for the analysis of pesticide residues in agricultural products. This paper summarizes recent research progress in the field from three main directions: sample pretreatment, SERS substrates, and data processing. The prospects and limitations of SERS technology are also discussed, in order to provide theoretical support for rapid detection of pesticide residues in Chinese herbal medicines.

Research on Defect Detection in Kubo Peach Based on Hyperspectral Imaging Technology Combined with CARS-MIV-GA-SVM Method.

Due to the dark red surface of ripe fresh peaches, their internal injury defects cannot be detected using the naked eye and conventional images. The rapid and accurate detection of fresh peach defects can improve the efficiency of fresh peach classification. The goal of this paper was to develop a nondestructive approach to simultaneously detecting internal injury defects and external injuries in fresh peaches. First, we collected spectral data from 347 Kubo peach samples using hyperspectral imaging technology (900-1700 nm) and carried out pretreatment. Four methods (the competitive adaptive reweighting algorithm (CARS), the combination of CARS and the average influence value algorithm (CARS-MIV), the combination of CARS and the successive projections algorithm (CARS-SPA), and the combination of CARS and uninformative variable elimination (CARS-UVE)) were used to extract the characteristic wavelength. Based on the characteristic wavelength extracted using the above methods, a genetic algorithm optimization support vector machine (GA-SVM) model and a least-squares support vector machine (LS-SVM) model were used to establish classification models. The results show that the combination of CARS and other feature wavelength extraction methods can effectively improve the prediction accuracy of the model when the number of wavelengths is small. Among them, the discriminant accuracy of the CARS-MIV-GA-SVM model reaches 93.15%. In summary, hyperspectral imaging technology can accomplish the accurate detection of Kubo peaches defects, and provides feasible ideas for the automatic classification of Kubo peaches.

Research on the Effects of Drying Temperature for the Detection of Soil Nitrogen by Near-Infrared Spectroscopy.

Nitrogen nitrates play a significant role in the soil's nutrient cycle, and near-infrared spectroscopy can efficiently and accurately detect the content of nitrate-nitrogen in the soil. Accordingly, it can provide a scientific basis for soil improvement and agricultural productivity by deeply examining the cycle and transformation pattern of nutrients in the soil. To investigate the impact of drying temperature on NIR soil nitrogen detection, soil samples with different N concentrations were dried at temperatures of 50 °C, 65 °C, 80 °C, and 95 °C, respectively. Additionally, soil samples naturally air-dried at room temperature (25 °C) were used as a control group. Different drying times were modified based on the drying temperature to completely eliminate the impact of moisture. Following data collection with an NIR spectrometer, the best preprocessing method was chosen to handle the raw data. Based on the feature bands chosen by the RFFS, CARS, and SPA methods, two linear models, PLSR and SVM, and a nonlinear ANN model were then established for analysis and comparison. It was found that the drying temperature had a great effect on the detection of soil nitrogen by near-infrared spectroscopy. In the meantime, the SPA-ANN model simultaneously yielded the best and most stable accuracy, with Rc2 = 0.998, Rp2 = 0.989, RMSEC = 0.178 g/kg, and RMSEP = 0.257 g/kg. The results showed that NIR spectroscopy had the least effect and the highest accuracy in detecting nitrogen at 80 °C soil drying temperature. This work provides a theoretical foundation for agricultural production in the future.

Estimation of rice seedling growth traits with an end-to-end multi-objective deep learning framework.

In recent years, rice seedling raising factories have gradually been promoted in China. The seedlings bred in the factory need to be selected manually and then transplanted to the field. Growth-related traits such as height and biomass are important indicators for quantifying the growth of rice seedlings. Nowadays, the development of image-based plant phenotyping has received increasing attention, however, there is still room for improvement in plant phenotyping methods to meet the demand for rapid, robust and low-cost extraction of phenotypic measurements from images in environmentally-controlled plant factories. In this study, a method based on convolutional neural networks (CNNs) and digital images was applied to estimate the growth of rice seedlings in a controlled environment. Specifically, an end-to-end framework consisting of hybrid CNNs took color images, scaling factor and image acquisition distance as input and directly predicted the shoot height (SH) and shoot fresh weight (SFW) after image segmentation. The results on the rice seedlings dataset collected by different optical sensors demonstrated that the proposed model outperformed compared random forest (RF) and regression CNN models (RCNN). The model achieved R2 values of 0.980 and 0.717, and normalized root mean square error (NRMSE) values of 2.64% and 17.23%, respectively. The hybrid CNNs method can learn the relationship between digital images and seedling growth traits, promising to provide a convenient and flexible estimation tool for the non-destructive monitoring of seedling growth in controlled environments.

Elucidating the Effect of Temperature Stress on the Protein Content, Total Antioxidant Capacity, and Antioxidant Enzyme Activities in Tetranychus urticae (Acari: Tetranychidae).

Tetranychus urticae Koch is a worldwide agricultural pest mite that feeds on more than 1100 kinds of crops. The mite has developed a high level of tolerance to high temperatures, but the physiological mechanism underlying the outstanding adaptability of this pest to high temperatures remains unclear. To clarify the physiological mechanisms of T. urticae in response to short-term heat stress, four temperatures (36, 39, 42, and 45 °C) and three short-term heat durations (2, 4, and 6 h) were conducted to test the effects on protein content, the activities of superoxide dismutase (SOD), peroxidase (POD), and catalase (CAT), and the total antioxidant capacity (T-AOC). The results showed that protein content, antioxidant enzyme activity, and T-AOC in T. urticae were significantly induced by heat stress. These results suggest that heat stress induces oxidative stress and that antioxidant enzymes play an important role in reducing oxidative damage in T. urticae. The data of this study will provide a basis for further research on the molecular mechanisms of thermostability and ecological adaptability of T. urticae.

A novel ratiometric fluorescent probe for sensitive detection of jasmonic acid in crops.

Herein, a novel ratiometric fluorescent probe was proposed for sensitive detection of jasmonic acid (JA) based on NCQDs@Co-MOFs@MIPs. The prepared NCQDs, with uniquely dual-emissive performance, are insensitive to JA due to electrostatic repulsion. Interestingly, the introduction of Co-MOFs not only avoided the self-aggregation of NCQDs, but changed the surface charge of NCQDs and triggered the response of NCQDs to JA. More importantly, the imprinted recognition sites from MIPs provided "key-lock" structures to specifically capture JA molecules, greatly improving the selectivity of the probe to JA. Under the synergistic actions of Co-MOFs and MIPs, JA can interact with NCQDs through photo-induced electron transfer (PET), resulting in the changes on emission intensity of the probe at Em = 367 nm and 442 nm. Based on the observations, the quantification of JA was realized in the range of 1-800 ng/mL with the limit of detection (LOD) of 0.35 ng/mL. In addition, the probe was used for detecting JA in rice with satisfactory analysis results, indicating the probe holds great potential for monitoring JA levels in crops. Overall, this strategy provides new insights into the construction of practical probes for sensitive detection of plant hormones in crops.

High-Precision Automatic Identification of Fentanyl-Related Drugs by Terahertz Spectroscopy with Molecular Dynamics Simulation and Spectral Similarity Mapping.

Fentanyl is a potent opioid analgesic with high bioavailability. It is the leading cause of drug addiction and overdose death. To better control the abuse of fentanyl and its derivatives, it is crucial to develop rapid and sensitive detection methods. However, fentanyl-related substrates undergo similar molecular structures resulting in similar properties, which are difficult to be identified by conventional spectroscopic methods. In this work, a method for the automatic identification of 8 fentanyl-related substances with similar spectral characteristics was developed using terahertz (THz) spectroscopy coupled with density functional theory (DFT) and spectral similarity mapping (SSM). To characterize the THz fingerprints of these fentanyl-related samples more accurately, the method of baseline estimation and denoising with sparsity was performed before revealing the unique molecular dynamics of each substance by DFT. The SSM method was proposed to identify these fentanyl analogs based on weighted spectral cosine-cross similarity and fingerprint discrete Fréchet distance, generating a matching list by stepwise searching the entire spectral database. The top matched list returned the identification results of the target fentanyl analogs with accuracies of 94.48~99.33%. Results from this work provide algorithms' increased reliability, which serves as an artificial intelligence-based tool for high-precision fentanyl analysis in real-world samples.

Rapid Detection of Available Nitrogen in Soil by Surface-Enhanced Raman Spectroscopy.

Soil-available nitrogen is the main nitrogen source that plants can directly absorb for assimilation. It is of great significance to detect the concentration of soil-available nitrogen in a simple, rapid and reliable method, which is beneficial to guiding agricultural production activities. This study confirmed that Raman spectroscopy is one such approach, especially after surface enhancement; its spectral response is more sensitive. Here, we collected three types of soils (chernozem, loess and laterite) and purchased two kinds of nitrogen fertilizers (ammonium sulfate and sodium nitrate) to determine ammonium nitrogen (NH4-N) and nitrate nitrogen (NO3-N) in the soil. The spectral data were acquired using a portable Raman spectrometer. Unique Raman characteristic peaks of NH4-N and NO3-N in different soils were found at 978 cm-1 and 1044 cm-1, respectively. Meanwhile, it was found that the enhancement of the Raman spectra by silver nanoparticles (AgNPs) was greater than that of gold nanoparticles (AuNPs). Combined with soil characteristics and nitrogen concentrations, Raman peak data were analyzed by multiple linear regression. The coefficient of determination for the validation (Rp2) of multiple linear regression prediction models for NH4-N and NO3-N were 0.976 and 0.937, respectively, which deeply interpreted the quantitative relationship among related physical quantities. Furthermore, all spectral data in the range of 400-2000 cm-1 were used to establish the partial least squares (PLS), back-propagation neural network (BPNN) and least squares support vector machine (LSSVM) models for quantification. After cross-validation and comparative analysis, the results showed that LSSVM optimized by particle swarm methodology had the highest accuracy and stability from an overall perspective. For all datasets of particle swarm optimization LSSVM (PSO-LSSVM), the Rp2 was above 0.99, the root mean square errors of prediction (RMSEP) were below 0.15, and the relative prediction deviation (RPD) was above 10. The ultra-portable Raman spectrometer, in combination with scatter-enhanced materials and machine learning algorithms, could be a promising solution for high-efficiency and real-time field detection of soil-available nitrogen.

Multimodal Bubble Microrobot Near an Air-Water Interface.

The development of multifunctional and robust swimming microrobots working at the free air-liquid interface has encountered challenge as new manipulation strategies are needed to overcome the complicated interfacial restrictions. Here, flexible but reliable mechanisms are shown that achieve a remote-control bubble microrobot with multiple working modes and high maneuverability by the assistance of a soft air-liquid interface. This bubble microrobot is developed from a hollow Janus microsphere (JM) regulated by a magnetic field, which can implement switchable working modes like pusher, gripper, anchor, and sweeper. The collapse of the microbubble and the accompanying directional jet flow play a key role for functioning in these working modes, which is analogous to a "bubble tentacle." Using a simple gamepad, the orientation and the navigation of the bubble microrobot can be easily manipulated. In particular, a speed modulation method is found for the bubble microrobot, which uses vertical magnetic field to control the orientation of the JM and the direction of the bubble-induced jet flow without changing the fuel concentration. The findings demonstrate a substantial advance of the bubble microrobot specifically working at the air-liquid interface and depict some nonintuitive mechanisms that can help develop more complicated microswimmers.

The Effect of Pressure-Shift Freezing versus Air Freezing and Liquid Immersion on the Quality of Frozen Fish during Storage.

In this study, a self-cooling laboratory system was used for pressure−shift freezing (PSF), and the effects of pressure−shift freezing (PSF) at 150 MPa on the quality of largemouth bass (Micropterus salmoides) during frozen storage at −30 °C were evaluated and compared with those of conventional air freezing (CAF) and liquid immersion freezing (LIF). The evaluated thawing loss and cooking loss of PSF were significantly lower than those of CAF and LIF during the whole frozen storage period. The thawing loss, L* value, b* value and TBARS of the frozen fish increased during the storage. After 28 days storage, the TBARS values of LIF and CAF were 0.54 and 0.65, respectively, significantly higher (p < 0.05) than the 0.25 observed for PSF. The pH of the samples showed a decreasing trend at first but then increased during the storage, and the CAF had the fastest increasing trend. Based on Raman spectra, the secondary structure of the protein in the PSF-treated samples was considered more stable. The α-helix content of the protein in the unfrozen sample was 59.3 ± 7.22, which decreased after 28 days of frozen storage for PSF, LIF and CAF to 48.5 ± 3.43, 39.1 ± 2.35 and 33.4 ± 4.21, respectively. The results showed that the quality of largemouth bass treated with PSF was better than LIT and CAF during the frozen storage.

Demonstration of Escherichia coli Inactivation in Sterile Physiological Saline under High Pressure (HP) Phase Transition Conditions and Analysis of Probable Contribution of HP Metastable Positions Using Model Solutions and Apple Juice.

It was demonstrated that the inactivation of high pressure (HP) treatment on Escherichia coli survival in sterile physiological saline (SPS) was influenced by the treatment conditions: unfrozen, frozen-thawed and fully frozen (phase transition). In order to probe the enhanced phase transition microbial destruction, vibration effects of phase transition position were created and discussed. Test samples were placed in HP chamber for treatment (150/240/330 MPa, no holding time) at room temperature and a special cooling device was used to maintain the phase transition conditions. Results showed that the phase transition from ice I to ice III of frozen SPS could be realized based on the cooling of a 20% sodium chloride solution. HP treatment under fully frozen conditions produced the best lethal effect compared to unfrozen and freeze-thaw samples. Vibration tests were carried out by using model solutions and apple juice to explore the behavior of phase transition. A synchronous and advance phase transition of internal apple juice was realized, respectively, by using pure water and 5% sodium chloride solution as external vibration sources, and the advance phase transitions of external pure water were realized by using 5% sodium chloride solution and 5% glucose solution as internal vibration sources.

Rapid Detection of Different Types of Soil Nitrogen Using Near-Infrared Hyperspectral Imaging.

Rapid and accurate determination of soil nitrogen supply capacity by detecting nitrogen content plays an important role in guiding agricultural production activities. In this study, near-infrared hyperspectral imaging (NIR-HSI) combined with two spectral preprocessing algorithms, two characteristic wavelength selection algorithms and two machine learning algorithms were applied to determine the content of soil nitrogen. Two types of soils (laterite and loess, collected in 2020) and three types of nitrogen fertilizers, namely, ammonium bicarbonate (ammonium nitrogen, NH4-N), sodium nitrate (nitrate nitrogen, NO3-N) and urea (urea nitrogen, urea-N), were studied. The NIR characteristic peaks of three types of nitrogen were assigned and regression models were established. By comparing the model average performance indexes after 100 runs, the best model suitable for the detection of nitrogen in different types was obtained. For NH4-N, R2p = 0.92, RMSEP = 0.77% and RPD = 3.63; for NO3-N, R2p = 0.92, RMSEP = 0.74% and RPD = 4.17; for urea-N, R2p = 0.96, RMSEP = 0.57% and RPD = 5.24. It can therefore be concluded that HSI spectroscopy combined with multivariate models is suitable for the high-precision detection of various soil N in soils. This study provided a research basis for the development of precision agriculture in the future.

Ultra-sensitive detection of hydrogen peroxide and levofloxacin using a dual-functional fluorescent probe.

Herein, a flower-shaped fluorescent probe was proposed for hydrogen peroxide (H2O2) and levofloxacin (LVF) sensing based on MoOx QDs@Co/Zn-MOFs with porous structure. Both MoOx QDs and Co/Zn-MOFs exhibited peroxidase-like properties, and the combination of them greatly aroused the synergistic catalytic capabilities between them. In o-Phenylenediamine (OPD)-H2O2 system, MoOx QDs@Co/Zn-MOFs efficiently catalyzed H2O2 to produce •OH and then oxidized OPD to its oxidation product (OxOPD). The OxOPD could not only emit blue fluorescence, but also inhibit the fluorescent intensity of MoOx QDs through fluorescence resonance energy transfer (FRET). Moreover, when introducing LVF into the system, the fluorescent intensities of MoOx QDs increased along with the aggregation of themselves while that of OxOPD remained unchanged, which was explained by the joint behavior of FRET and photo-induced electron transfer (PET) instead of the conventional aggregation-induced emission enhancement (AIEE). With these observation, the proposed probe was employed for H2O2 and LVF determination in biological samples with the limit of detection (LOD) of 32.60 pmol/L and 0.85 µmol/L, respectively, suggesting the method holds great promises for trace H2O2 and LVF monitoring in eco-environment.

Structure analysis and non-invasive detection of cadmium-phytochelatin2 complexes in plant by deep learning Raman spectrum.

Plants synthesize phytochelatins to chelate in vivo toxic heavy metal ions and produce nontoxic complexes for tolerating the stress. Detection of the complexes would simplify the identification of high phytoremediation cultivars, as well as assessment of plant food for safe consumption. Thus, a confocal Raman spectroscopy combined with density functional theory and deep learning was used for characterizing phytochelatin2 (PC2), and Cd-PC2 mixtures. Results showed the PC2 chelate Cd2+ in a 2:1 ratio to produce Cd(PC2)2; Cd-S bonds of the Cd(PC2)2 have signature Raman vibrations at 305 and 610 cm-1 which are the most distinctive spectral signatures for varieties of Cd-PCs complexes. The PC2 was used as a natural probe to stabilize the chemical status of Cd, and to enrich and magnify Raman signature of the trace Cd for deep learning models which enabled condition of the Cd(PC2)2 in pak choi leaf to be visualized, quantified, and classified by directly using raw spectra of the leaf. This study provides a general protocol by using Raman information for structure analysis and non-invasive detection of heavy metal-PCs complexes in plants and provides a novel idea for simplifying identification of high phytoremediation cultivars, as well as assessment of heavy metal related food safeties.

Analyzing cadmium-phytochelatin2 complexes in plant using terahertz and circular dichroism information.

Phytochelatins are plants' small metal-binding peptides which chelate internal heavy metals to form nontoxic complexes. Detecting the complexes in plants would simplify identification of cultivars with both high tolerance and enrichment capabilities for heavy metals which represent phytoextraction performance. Thus, a terahertz spectroscopy combined with density functional theory, chemometrics and circular dichroism was used for characterization of phytochelatin2 (PC2), Cd-PC2 mixture standards, and pak choi (Brassica chinensis) leaves as a plant model. Results showed PC2 chelates Cd2+ in a 2:1 ratio to form Cd(PC2)2 complex; Cd connected to thoils of PC2 and changed ß-turn and random coil of PC2 peptide chain to ß-Sheet which presented as terahertz vibrations of PC2 around 1.03 and 1.71 THz being suppressed; the best models for detecting the complex in pak choi were obtained by partial least squares regression modeling combined with successive projections algorithm selection; the models used PC2 as a natural probe for visualizing and quantifying chelated Cd in pak choi leaf and achieved a limit of detection up to 1.151 ppm. This study suggested that terahertz information of the heavy metal-PCs complexes is qualified for representing a simpler alternative to classical index for evaluating phytoextraction performance of plant; it provided a general protocol for structure analysis and detection of heavy metal-PCs complexes in plant by terahertz absorbance.

Terahertz dual-band metamaterial absorber for trace indole-3-acetic acid and tricyclazole molecular detection based on spectral response analysis.

A terahertz (THz) metamaterial absorber based on a split ring resonator (SRR) structure was used to realize the highly accurate detection of trace pesticides, including indole-3-acetic acid (IAA) and tricyclazole. The density functional theory (DFT) was used to analysis the THz fingerprint peaks of IAA and tricyclazole. According to the dual-band (0.918 and 1.575 THz) near-perfect absorption characteristics of the absorber in the transverse magnetic (TM) polarization state, the univariate regressions were used to analyze the responses of peak amplitude and frequency to pesticide concentrations. For IAA, the sensing response based on the peak amplitude at 1.575 THz was the best with a coefficient of determination (R2) of 0.9627. As for tricyclazole, the best sensing response was based on the peak frequency at 1.575 THz with a R2 of 0.8742. Moreover, the detection accuracy of IAA (R2 = 0.9752) and tricyclazole (R2 = 0.9177) were significantly improved through effective variable selection and multivariate fusion. The results indicated that the limit of detection (LOD) of two pesticides both reached 10 ng/L. This study provided a good experimental basis for trace hazardous substances detection, presenting a new prospect for food quality and safety control in the future.

Gold nanoparticles-mediated ratiometric fluorescence aptasensor for ultra-sensitive detection of Abscisic Acid.

Herein, a novel ratiometric aptasensor based on carbon quantum dots@2-Methylimidazole zinc salt (CQDs@ZIF-8) and aptamer-functionalized gold nanoparticles (Apt-AuNPs) was developed for highly sensitive detection of ABA by fluorescence spectrometry. The CQDs@ZIF-8 nanomaterials displayed dual-emission properties at 490 nm and 657 nm with excitation at 420 nm were synthesized for the first time. ZIF-8 not only served as an anchor point for CQDs but also acted as a modulator to regulate fluorescence signals of CQDs. Interestingly, introduction of ZIF-8 changed the quenching properties of the AuNPs on CQDs. The AuNPs quenched the fluorescence of CQDs@ZIF-8 at 490 nm but not at the second peak of 657 nm. Few studies have been reported on the ineffectiveness of AuNPs in fluorescence quenching as far as we know. In this study, we found that incorporation of ABA triggered the aggregation of AuNPs due to the specific ABA-aptamer recognition and this changed the fluorescence intensity of the ratiometric probe (CQDs@ZIF-8@Apt-AuNPs). The proposed probe increased the sensitivity and selectivity of determining ABA levels in rice seeds in the range of 0.100-150 ng/mL with an LOD of 30.0 ng/L. Importantly, the method proposed here offers a new unique strategy for the construction of ratiometric probes and ultra-sensitive measurement of biomolecules.

Trace Identification and Visualization of Multiple Benzimidazole Pesticide Residues on Toona sinensis Leaves Using Terahertz Imaging Combined with Deep Learning.

Molecular spectroscopy has been widely used to identify pesticides. The main limitation of this approach is the difficulty of identifying pesticides with similar molecular structures. When these pesticide residues are in trace and mixed states in plants, it poses great challenges for practical identification. This study proposed a state-of-the-art method for the rapid identification of trace (10 mg·L-1) and multiple similar benzimidazole pesticide residues on the surface of Toona sinensis leaves, mainly including benzoyl (BNL), carbendazim (BCM), thiabendazole (TBZ), and their mixtures. The new method combines high-throughput terahertz (THz) imaging technology with a deep learning framework. To further improve the model reliability beyond the THz fingerprint peaks (BNL: 0.70, 1.07, 2.20 THz; BCM: 1.16, 1.35, 2.32 THz; TBZ: 0.92, 1.24, 1.66, 1.95, 2.58 THz), we extracted the absorption spectra in frequencies of 0.2-2.2 THz from images as the input to the deep convolution neural network (DCNN). Compared with fuzzy Sammon clustering and four back-propagation neural network (BPNN) models (TrainCGB, TrainCGF, TrainCGP, and TrainRP), DCNN achieved the highest prediction accuracies of 100%, 94.51%, 96.26%, 94.64%, 98.81%, 94.90%, 96.17%, and 96.99% for the control check group, BNL, BCM, TBZ, BNL + BCM, BNL + TBZ, BCM + TBZ, and BNL + BCM + TBZ, respectively. Taking advantage of THz imaging and DCNN, the image visualization of pesticide distribution and residue types on leaves was realized simultaneously. The results demonstrated that THz imaging and deep learning can be potentially adopted for rapid-sensing detection of trace multi-residues on leaf surfaces, which is of great significance for agriculture and food safety.