JumpDetector: An automated monitoring equipment for the locomotion of jumping insects

Abstract

Continuous jumping behavior, a kind of endurance locomotion, plays important roles in insect ecological adaption and survival. However, the methods used for the efficient evaluation of insect jumping behavior remain largely lacking. Here, we developed a locomotion detection system named JumpDetector with automatic trajectory tracking and data analysis to evaluate the jumping of insects. This automated system exhibits more accurate, efficient, and adjustable performance than manual methods. By using this automatic system, we characterized a gradually declining pattern of continuous jumping behavior in 4th-instar nymphs of the migratory locust. We found that locusts in their gregarious phase outperformed locusts in their solitary phase in the endurance jumping locomotion. Therefore, the JumpDetector could be widely used in jumping behavior and endurance locomotion measurement.

Keywords: EthoVision; Locusta migratoria; bio-artificial interaction; endurance locomotion; jumping behavior.

Figure 1

Schematic and detection method of JumpDetector. (A) Image of JumpDetector. (B) Structural composition of a forced device. (C) Forced stimulus intensities provided by our detection platform and defined as forced ranks from I to IV. (D) Tracking and zone regionalization of jumping trajectory, (a): trajectory tracking (1: upward, 2: peak, 3: downward), (b): zone regionalization of jumping trajectory (yellow circle represents the peak zone; white circle displays the intermediate zones). The red line indicates the jumping trajectory, and the red spot represents the center of the recognized target). (E) Definition of locomotion parameters in a jumping event (the frequency in the peak zone (peak frequency) indicates the frequency of movement, and the height of the peak zone shows the distance moved). (F) Flowchart demonstrating the data recording process by using an automatic script J‐Recorder.

Figure 2

Accuracy and efficiency of the automatic video tracking and data recording methods. (A) Comparison of TFM and TDM between the automatic method and the ground truth for each locust in 1 min movies (n = 30). (B) Comparison of the efficiency between the automatic and manual operations. TFM, total frequency of movement; TDM, total distance moved. *indicates a significant difference between automatic method and manual operation (Student's t‐test, ***P < 0.001).

Figure 3

Performance evaluation of JumpDetector. (A) TFM and TDM significantly increased as the forced rank increased from I to III and stopped increasing at forced rank IV (within 10 min forced movement). NO indicates group without forced conditions. (B) Activity thermography analysis showed the increasing tendency of the moving area and the increasing forced rank from I to III. (C) TFM and TDM in the forced group significantly increased over time (30 min) compared with those of the control group. Significant differences at different points are denoted by letters with different colors (n = 31 to 36 animals per point, one‐way analysis of variance (ANOVA), P < 0.05). *indicates significant difference between “Control” (without forced conditions) and “Forced” (with forced conditions) group locusts (Student's t‐test, ***P < 0.001). TFM, total frequency of movement; TDM, total distance moved.

Figure 4

Characterization of the endurance locomotion of 4th‐instar locust nymphs. (A) Dynamic patterns of TFM and TDM gradually declined after the first time interval under different window sizes at the 15, 10, and 5 min time intervals. (B) Dynamic patterns of TFM and TDM gradually decreased with drastic (stage I: 0–10 min), continued (stage II: 10–30 min), and slow (stage III: 30–60 min) decline under the window size at 1 min. TFM, total frequency of movement; TDM, total distance moved.

Figure 5

Different performances of endurance locomotion between gregarious and solitary locusts. (A–B) Both TFM and TDM values displayed a gradually decreasing pattern in the two locust phases within 1 h forced movement. Both values were significantly higher in gregarious locusts than in solitary locusts at each time interval or total time. Repeated measures analysis of variance (ANOVA) is used to compare the locomotor abilities between the gregarious locusts and the solitary locusts at different time intervals (n = 42 animals per point, difference is considered significant at P < 0.05). *indicates a significant difference between solitary and gregarious locusts (n = 42 animals per group, Student's t‐test, **P < 0.01, ***P < 0.001). TFM, total frequency of movement; TDM, total distance moved.