TWI876701B - Industrial safety warning system and method - Google Patents

Abstract

An industrial safety warning system includes a camera device and a computing host. The camera device captures images facing a detection area to generate a plurality of captured frames. The computing host is communicably connected to the camera device for retrieving the plurality of captured frames, and selectively issues a warning message based on the plurality of captured frames. In particular, selectively issuing the warning message by the computing host includes: detecting, by the computing host, whether a portion corresponding to the detection area in the plurality of captured frames has a human body image; positioning, by the computing host, a plurality of extremities based on the limbs of the human body image when the portion has a human body image; and issuing, by the computing host, the warning message of the first type when any of the extremity positions remains unchanged for a predetermined duration.

Description

Industrial safety warning system and method

The present invention relates to a display screen driving method and a display screen driving device.

In high potential hazardous sites, such as petrochemical plants, pharmaceutical plants, fuel storage depots, etc., it is common to see multiple large vehicles driving to designated locations to load or unload hazardous materials. There will also be staff at designated locations responsible for pre-processing before unloading, or post-processing after the unloading is completed and the vehicles drive away.

However, drivers of large vehicles may not always maintain a high level of alertness and may unload goods while pre-processing is still in progress, resulting in workplace safety risks for on-site personnel. On-site personnel may also start subsequent operations before the large vehicle leaves the designated location due to their own lack of concentration or lack of workplace safety awareness, thereby increasing unnecessary workplace safety risks. The above-mentioned workplace safety hazards include, for example, explosions, fires, and the release of toxic gases. In order to prevent these hazards from occurring, effective safety measures need to be taken.

In view of the above, the present invention provides a work safety warning system and method, which reduces unnecessary work safety risks through real-time monitoring mode combined with different types of alarm mechanisms.

According to an embodiment of the present invention, a work safety warning system includes a camera and a computing host. The camera captures images toward a detection area to generate multiple image frames. The computing host is communicatively connected to the camera, and the computing host is used to obtain the multiple image frames and selectively issue warning messages based on these image frames. Among them, the computing host selectively issues warning messages including: detecting whether a part of the multiple image frames corresponding to the detection area has a human body image, when the part has a human body image, locating multiple end positions according to the limbs in the human body image, and issuing a first type of warning message when any end position has not changed for a preset period of time.

According to an embodiment of the present invention, a method for industrial safety warning includes: a camera captures images toward a detection area to generate a plurality of image frames, a computing host connected to the camera captures the plurality of image frames, and selectively issues a warning message based on the image frames. The computing host selectively issues a warning message including: detecting whether a portion of the plurality of image frames corresponding to the detection area has a human image, and when the portion has a human image, locating a plurality of end positions based on the limbs in the human image, and issuing a first type of warning message when any end position does not change for a preset period of time.

In summary, the industrial safety warning system and method proposed by the present invention can continuously monitor the dangerous factors that may occur during the operation process. When the limbs of the on-site operator are stuck and unable to move, an on-site alarm is immediately issued to notify all personnel on site, thereby avoiding potential industrial safety accidents.

The above description of the disclosed content and the following description of the implementation methods are used to demonstrate and explain the spirit and principle of the present invention, and provide a further explanation of the scope of the patent application of the present invention.

The following detailed description of the features and advantages of the present invention is provided in the implementation mode, and the content is sufficient to enable any person skilled in the relevant art to understand the technical content of the present invention and implement it accordingly. According to the content disclosed in this specification, the scope of the patent application and the drawings, any person skilled in the relevant art can easily understand the relevant purposes and advantages of the present invention. The following embodiments are to further illustrate the viewpoints of the present invention, but are not to limit the scope of the present invention by any viewpoint.

FIG. 1 is a block diagram of a work safety warning system according to the first embodiment of the present invention. As shown in FIG. 1 , the work safety warning system 10 includes a camera 1 and a computing host 3. FIG. 2 is a block diagram of a work safety warning system according to the second embodiment of the present invention. Compared with the work safety warning system 10 of the first embodiment, the work safety warning system 20 of the second embodiment further includes a vehicle sensor 5. FIG. 3 is a schematic diagram of an application scenario example of the work safety warning system 20 of the second embodiment of the present invention.

Please refer to Figures 1 to 3. The camera 1 captures images toward a detection area A1 to generate a plurality of captured images.

The computing host 3 is communicatively connected to the photographing device 1. The computing host 3 is used to obtain a plurality of images taken by the photographing device 1, and then selectively issues a warning message based on these images. In one embodiment, the computing host 3 is, for example, a personal computer or a server, and the present invention is not limited thereto.

The computing host 3 is used to detect whether a portion of the plurality of image capture images corresponding to the detection area A1 has a human body image. If it is determined to be yes, multiple end positions are located according to the limbs H1, H2, H3, and H4 in the human body image. When any end position remains unchanged for a preset time (e.g., 5 minutes), the computing host 3 issues a first type of warning message. The first type of warning message is, for example, notified to the control center of the factory for subsequent processing.

The vehicle sensor 5 is communicatively connected to the computing host 3. The vehicle sensor 5 is used to sense a moving vehicle T in a driving area A2, wherein the driving area A2 includes the detection area A1. In one embodiment, the vehicle sensor 5 can be a weight sensor, a camera, or an infrared sensor, but the present invention is not limited thereto. When the computing host 3 detects a human image and a moving vehicle T at the same time, it issues a second type of warning message. The second type of warning message is, for example, a warning sound or a warning light signal on the spot.

FIG. 4 is a flow chart of a work safety warning method according to an embodiment of the present invention.

As shown in FIG4 , the method includes steps S1 and S3. In step S1, the camera 1 takes images toward the detection area A1 to generate a plurality of image frames. In step S3, the computing host 3 obtains the plurality of image frames taken by the camera 1 and selectively issues a warning message based on the image frames.

Fig. 5 is a flow chart of the first implementation method of step S3 in Fig. 4. As shown in Fig. 5, the first implementation method includes steps S31 to S34. The first implementation method is applicable to the industrial safety warning system 10 of the first embodiment.

Step S31 follows step S1. In step S31, the computing host 3 determines whether the portion corresponding to the detection area A1 in the plurality of image-taking frames has a human body image. If the determination is yes, the computing host 3 proceeds to step S32; if the determination is no, the computing host 3 returns to step S1 and continues to shoot more image-taking frames.

In one embodiment, the computing host 3 can use OpenCV with a human body feature model to detect human body images in the image capture frame. The present invention does not limit the human body feature model or the algorithm used by the computing host to detect human body images.

In step S32, the computing host 3 locates multiple end positions according to the limbs H1, H2, H3, and H4 in the human body image. The multiple image frames taken in time sequence are _F1 , _F2 , ... _Fi , and _FN . Assuming that the computing host 3 detects the human body image in the i-th image frame _Fi , and locates the four end positions corresponding to the limbs, the computing host 3 will continue to confirm whether the previously located end position changes from the original position in the image frames Fi ₊₁ , Fi ₊₂ , ..., _FN. By determining whether the location of the end position changes, it can be confirmed whether the staff in the detection area A1 is in a normal state. In a normal state, the staff will not remain motionless for more than a period of time.

In step S33, if any end position does not change for a preset time (e.g., 5 minutes), it means that the personnel in the detection area may have their hands or feet stuck in the factory equipment due to some factors. If the judgment of step S33 is yes, then continue to step S34; if the judgment of step S33 is no, then return to step S32. In an embodiment of step S33, the computing host 3 sets a timer for each end position and resets the timer when the end position changes.

In step S34, the computing host 3 issues a first type of warning message. In one embodiment, the warning message includes a first type and a second type. The first type of warning message is, for example, in the form of a text notification, which is used to feed back to the control center. The second type of warning message is, for example, a warning sound or a warning light, which is used to broadcast in the field around the detection area to notify the on-site personnel that a work safety hazard is about to occur or has occurred.

Fig. 6 is a flow chart of the second implementation of step S3 in Fig. 4. As shown in Fig. 6, the second implementation includes steps S31 to S35 and S38. The second implementation is applicable to the industrial safety warning system 20 of the second embodiment.

Step S35 is performed when the judgment of step S1 is yes. In step S35, the vehicle sensor 5 is used to sense the moving vehicle T in the driving area A2, and the computing host 3 confirms whether there is a moving vehicle T in the driving area A2 according to the sensing result of the vehicle sensor 5. If the judgments of step S35 and step S31 are both yes, in other words, if the human body image and the moving vehicle T are detected at the same time, then continue to step S38. If the judgment of step S35 is no, then return to step S1 and continue to shoot more image frames.

In step S38, the computing host 3 issues a second type of warning message. The second type of warning message is, for example, a warning sound or a warning light, which is broadcast in the field surrounding the detection area to inform on-site personnel that a work safety hazard is about to occur or has occurred.

Fig. 7 is a flow chart of the third implementation of step S3 in Fig. 4. As shown in Fig. 7, the third implementation includes steps S31 to S34, S36 and S38. The third implementation is applicable to the industrial safety warning system 10 of the first embodiment.

Step S36 is executed between step S31 and step S32. In other words, after the computing host 3 detects the human body image and before locating the multiple end positions, the computing host 3 checks the current time and the work schedule. If the current time falls within a feeding time period of the work schedule, step S36 determines that it is yes, and continues to step S38 to issue a second type of warning message. On the contrary, if the current time does not fall within the feeding time period of the work schedule, then continue to step S32 and the subsequent processes.

Fig. 8 is a flow chart of the fourth embodiment of step S3 in Fig. 4. As shown in Fig. 8, the fourth embodiment includes steps S31 to S34, S37 and S38. The fourth embodiment is applicable to the industrial safety warning system 10 of the first embodiment.

Step S37 is executed after step S32. In other words, after the computing host 3 locates multiple end positions according to the limbs in the human body image, the computing host also calculates the change speed of each end position. If the change speed of any end position is greater than the preset value, the judgment of step S37 is yes, and step S38 is continued to issue a second type of warning message. On the contrary, if the change speed of all end positions is not greater than the preset value, it means that the judgment of step S37 is no, so the process after step S33 is continued.

In summary, the industrial safety warning system and method proposed by the present invention can continuously monitor the dangerous factors that may occur during the operation process. When the limbs of the on-site operator are stuck and unable to move, an on-site alarm is immediately issued to notify all personnel on site, thereby avoiding potential industrial safety accidents.

Although the present invention is disclosed as above with the aforementioned embodiments, it is not intended to limit the present invention. Any changes and modifications made within the spirit and scope of the present invention are within the scope of patent protection of the present invention. Please refer to the attached patent application for the scope of protection defined by the present invention.

1: Camera device 10,20: Industrial safety warning system 3: Computing host 5: Vehicle sensor A1: Detection area A2: Driving area H1,H2,H3,H4: Limbs T: Moving vehicle S1,S3,S31-S38: Steps

FIG1 is a block diagram of a work safety warning system according to the first embodiment of the present invention; FIG2 is a block diagram of a work safety warning system according to the second embodiment of the present invention; FIG3 is a schematic diagram of an application scenario example of the work safety warning system according to the second embodiment of the present invention; FIG4 is a flow chart of a work safety warning method according to an embodiment of the present invention; FIG5 is a flow chart of a first embodiment of a step in FIG4; FIG6 is a flow chart of a second embodiment of a step in FIG4; FIG7 is a flow chart of a third embodiment of a step in FIG4; and FIG8 is a flow chart of a fourth embodiment of a step in FIG4.

S1, S3: Steps

Claims (6)

A work safety warning system, comprising: A camera, which takes images toward a detection area to generate multiple image frames; and A computing host, which is communicatively connected to the camera, and is used to obtain the image frames and selectively issue a warning message based on the image frames; Wherein, the selective issuance of the warning message by the computing host includes: Detecting whether a portion of the image frames corresponding to the detection area has a human body image; When the portion has the human body image, locating multiple end positions based on the limbs in the human body image; When any of the end positions remains unchanged for a preset period of time, issuing the first type of the warning message; After the computing host locates the end positions according to the limbs in the human body image, the change speed of the position of any one of the end positions is calculated; and When the change speed is greater than a preset value, the second type of warning message is issued. The industrial safety warning system as described in claim 1 further includes: A vehicle sensor, communicatively connected to the computing host, the vehicle sensor is used to sense a moving vehicle in a driving area, wherein the driving area includes the detection area; and The selective issuance of the warning message by the computing host further includes: when the human body image and the moving vehicle are detected simultaneously, issuing the second type of the warning message. The industrial safety warning system as described in claim 1, wherein the selective issuance of the warning message by the computing host further includes: After the computing host detects the human body image and before locating the end positions, the computing host checks a current time and a work schedule; and When the current time falls within a feeding time period of the work schedule, the second type of the warning message is issued. A method for industrial safety warning, comprising: Using a camera to capture images of a detection area to generate multiple image frames; Using a computing host connected to the camera to obtain the image frames, and selectively issuing a warning message based on the image frames; Wherein, the selective issuance of the warning message by the computing host includes: Detecting whether a portion of the image frames corresponding to the detection area has a human body image; When the portion has the human body image, locating multiple end positions based on the limbs in the human body image; When any of the end positions does not change for a preset time, issuing the first type of the warning message; After the computing host locates the end positions based on the limbs in the human body image, calculating the change speed of the position of any of the end positions; and When the change rate is greater than a preset value, the second type of warning message is issued. The industrial safety warning method as described in claim 4 further includes: Using a vehicle sensor that is communicatively connected to the computing host to sense a moving vehicle in a driving area, wherein the driving area includes the detection area; and Using the computing host to selectively issue the warning message further includes: when the human body image and the moving vehicle are detected simultaneously, issuing the second type of the warning message. The industrial safety warning method as described in claim 4, wherein the selective issuance of the warning message by the computing host further includes: After the computing host detects the human body image and before locating the end positions, the computing host checks a current time and a work schedule; and When the current time falls within a feeding time period of the work schedule, the second type of the warning message is issued.

Images