Abstract: This study aims to detect salmon freshness using intelligent sensing with the dual-channel fluorescent probes. An intelligent sensing was developed using a dual-mode responsive fluorescent probe. The highly sensitive detection and visual tracking of biogenic amines (BAs) were realized during fish spoilage. 7-(diethylamino) coumarin-3-carbaldehyde (DEAC) was used as the fluorescent core, while the 1,3-dimethylbarbituric acid (DA) was introduced as the specific recognition unit. A fluorescent probe DEAC-DA was then achieved in the dual-channel colorimetric-fluorescent response to the BAs. A sensor label (SL) DEAC-DA/PVASL was fabricated to integrate with the mobile APP (Visual Evaluation) using polyvinyl alcohol (PVA) film technology. The freshness detection of salmon was applied to determine the high linear correlation between the red green blue (RGB) of the sensor label and the total volatile basic nitrogen (TVB-N) content of the salmon meat. A series of tests was carried out to verify the performance of the probes. The results showed that the DEAC-DA in the CH₃CN/H₂O (the volume ratio is 3:7) solution exhibited a dual-channel response of the color from pink to yellow and fluorescence from no fluorescence to bright blue for 15 types of BAs. The excellent performance was achieved in the rapid response (8 s), detection limit = 4.669 3 μmol/L), and broad range (0~400 μmol/L, R²=0.999 6). High-resolution mass spectrometry (HRMS) analysis and quantum chemical calculations were employed to clarify the sensing mechanism of the DEAC-DA to BAs. Thereby, the reaction mechanism was determined at the molecular level. The DEAC-DA experienced the fluorescence quenching due to the photoinduced electron transfer (PET). The fluorescence was regained during radiative transition under light excitation, particularly after nucleophilic addition with the BAs, thereby indicating an "open" type fluorescence response. According to the excellent recognition performance of the probe, three substrates (filter study, agarose, and PVA) were immobilized to fabricate three sensing labels. Ultimately, the sensing label DEAC-DA/PVASL was selected, according to the response after exposure to various amine vapors. This sensor label also exhibited the dual-channel visual response to the multiple BAs, indicating the high stability at 4 °C. There was a variation in the TVB-N, thiobarbituric acid (TBA), total viable count (TVC), and pH values. The sensing label was also employed to quantitatively monitor the freshness of salmon stored at 4 °C for 0~8 d. As the freshness of salmon changed from fresh to sub-fresh and then to spoiled, the DEAC-DA/PVASL changed from yellowish green to yellow and then to light yellow under daylight, while from emerald green to blue-green and then to bright blue under ultraviolet light. The software was also developed for the intelligent sensing system. In the application of the real-time monitoring of the salmon freshness, the RGB values of the sensing label were automatically extracted to obtain the linear relationship (R²=0.999) between (R+G)/(R+G+B) and the TVB-N of salmon. Additionally, this intelligent sensing system was also employed to monitor the freshness of the salmon stored at 25 °C. The sensing labels of different freshness levels were randomly selected to scan using the APP. The readings were converted into the TVB-N values. The TVB-N values were then compared with the Chinese national standard. There were small differences and high consistency between the two datasets. The monitoring accuracy was validated for the intelligent sensing system, indicating the high reliability of the detection. To summarize, the detection system was developed with a dual pattern and intelligent visualization. This finding can provide an efficient and economical solution for the quality monitoring of the aquatic products in cold chain logistics, in order to promote food preservation towards intelligent and on-site direction.