TW201928828A - Customer dynamic tracking and method thereof - Google Patents

Abstract

A customer dynamic tracking and method thereof is disclosed. By sensing a facial feature, a voiceprint feature, a weight data and a shoe feature as a customer message, and generating a shopping path and a stay time according to the shoe feature and the weight data through a plurality of sensing floors, and receiving a voice message to perform voiceprint and semantic analysis to obtain a corresponding shopping preference, so as to embed the shopping preferences, the shopping path and the stay time into the customer message for providing businesses as an adjusting basis of purchase and sales strategy. The mechanism is help to achieve the unused radio frequency identification (RFID) in the premise, tracking customer dynamics and shopping preferences.

Description

Customer dynamic tracking system and method thereof

The present invention relates to a tracking system and method thereof, and more particularly to a customer dynamic tracking system and method thereof that can track customer dynamics and learn shopping preferences without using radio frequency identification technology.

In recent years, with the rise of unmanned stores, how to track the dynamics of customers in the store, such as: picking up goods, returning goods, shopping routes, product attention time, etc., to judge the customer's shopping habits and even preferences, has become The problems that various manufacturers are eager to solve.

In general, tracking the dynamics of customers in the store can be achieved through many camera components. This method can only know the customer's mobile route, it is difficult to obtain the customer's shopping preferences, and there may be camera dead angles or because there are too many camera components. It causes the customer's shopping pressure and so on, so it is difficult to track the customer's dynamics and know the shopping preferences.

In view of this, some manufacturers have proposed a method of radio frequency identification (RFID), which sets a corresponding RFID tag (RFID Tag) on each item for identification by RFID tag at checkout. The product, in turn, knows the customer's shopping preferences. However, although this method can know the customer's shopping preferences, it takes a lot of cost to install the RFID tag on each product, and only the goods that the customer finally purchases are known. It is impossible to know the customer's dynamics and to know the products that the customers like but hesitate according to the customer's dynamics. Therefore, it is still impossible to effectively solve the problem that it is difficult to track customer dynamics and learn about shopping preferences.

In summary, it can be seen that in the prior art, it has been a long time that it is difficult to track customer dynamics and to know the shopping preferences. Therefore, it is necessary to propose improved technical means to solve this problem.

The invention discloses a customer dynamic tracking system and a method thereof.

First, the present invention discloses a customer dynamic tracking system, which includes: a customer database, a sensing module, a path module, a voice recognition module, and a processing module. The customer database is used for storing customer information; the sensing module includes an entrance sensor and is disposed at the entrance gate for sensing and recording facial features, voiceprint characteristics, and weight when the person enters the entrance gate. The data and the shoe print feature are stored in the customer database as a customer message corresponding to the person; the path module is used to electrically connect the sensing floor, and each sensor floor is used to sense the shoe print feature and the weight data, and continues to be sensed according to the sensed The shoe print feature and the weight data generate a corresponding shopping path and dwell time; the voice recognition module is electrically connected to the radio device for receiving the corresponding voice message through each of the radio devices, and performing voiceprint and semantic analysis on the voice message. Obtaining a shopping preference corresponding to the voiceprint feature; the processing module is configured to embed the shopping preferences of the corresponding voiceprint feature into the customer message having the same voiceprint feature, and the shopping path generated according to the sensed shoeprint feature and the weight data And the time of stay, embedded in the customer message with the same shoe print features and weight data.

In addition, the present invention discloses a customer dynamic tracking method, the steps of which include: providing an inductive floor, a sound pickup device, and an entrance gate, wherein the entrance gate includes an entrance sensor; the entrance sensor senses when a person enters the entrance gate Record facial features, voiceprint features, weight data and shoe print features for storage in the customer database as customer information for the person; the sensor floor continuously senses the shoe print characteristics and weight data, and continues to sense the shoe print features and The weight data generates a corresponding shopping path and a staying time; the corresponding voice message is received through the sound receiving device, and the voice message is subjected to voiceprint and semantic analysis to obtain a shopping preference corresponding to the voiceprint feature; It is preferred to embed the customer message having the same voiceprint feature, and to insert the shopping route and the stay time generated based on the sensed shoe print feature and the weight data into the customer message having the same shoe print feature and weight data.

The system and method disclosed by the present invention are as above, and differ from the prior art in that the present invention transmits a facial print, a voiceprint feature, a weight data, and a shoe print feature as customer information, and continues to sense the shoe print based on the sensing floor. Feature and weight data generate corresponding shopping paths and stay times, and receive voice messages for voiceprint and semantic analysis to obtain corresponding shopping preferences, so as to embed shopping preferences, shopping paths and staying time into customer messages, providing merchants as purchases and The basis for the adjustment of the sales strategy is to achieve tracking of customer dynamics and knowledge of shopping preferences without using RFID technology.

Through the above technical means, the present invention can achieve the technical effect of tracking customer dynamics and knowing shopping preferences without using radio frequency identification technology.

The embodiments of the present invention will be described in detail below with reference to the drawings and embodiments, so that the application of the technical means to solve the technical problems and achieve the technical effects can be fully understood and implemented.

Before describing the customer dynamic tracking system and method thereof disclosed in the present invention, the nouns defined by the present invention are first described. The shopping path according to the present invention refers to the sensing according to the customer (or called a person). The path created by the floor, in actual implementation, even if the shoe print and the same weight, but because each person walks different postures and points of force are different, so the sensed shoe print features are not the same. Therefore, according to the shoe print feature and the weight data, detecting the sensor floor that is sequentially stepped on can generate a corresponding path as a shopping path.

The customer dynamic tracking system and the method thereof are further described below with reference to the drawings. Please refer to "FIG. 1", "FIG. 1" is a system block diagram of the customer dynamic tracking system of the present invention, and the system includes: a customer database. 110. The sensing module 120, the path module 130, the voice recognition module 140, and the processing module 150. The customer database 110 is used to store customer information. In actual implementation, the customer database 110 can be a relational database using a Structured Query Language (SQL), a database that does not use SQL (NoSQL), or a text file directly as a database. The carrier is implemented in a manner such as a carrier.

The sensing module 120 includes various inlet sensors and is disposed at the entrance gate for sensing and recording facial features, voiceprint features, weight data, and shoe prints when a person enters the entrance gate. Features are stored in the customer database as customer information for this person. For example, a sensor-sensing device (CCD) is used to sense facial features, and a sound sensor (or a sound pickup device) senses the voice of the person to generate a voiceprint feature and a weight sensor sense. The weight data of the person is measured, and the sensing floor senses the shoe print characteristics of the person.

The path module 130 is configured to electrically connect all the sensing floors, each sensing floor is used to sense the shoe printing features and the weight data, and continuously generate corresponding shopping paths and residence time according to the sensed shoe printing features and weight data. In an actual implementation, the sensing floor comprises a silicone layer, an acryl layer, a photosensitive coupling element, and an infrared light emitting diode disposed on a side of the sensing floor for pressing the silicone layer and contacting the acrylic layer. At the time, the infrared rays propagating in the acrylic layer are scattered, and the scattered infrared rays are extracted by the photosensitive coupling elements to generate a shoe print feature. In addition, the sensing floor further includes a pressure sensing layer for sensing the weight data of the person.

The voice recognition module 140 is electrically connected to the sound receiving device for receiving the corresponding voice message through each of the sound receiving devices, and performing voiceprint and semantic analysis on the voice message to obtain shopping preferences corresponding to the voiceprint feature. In practical implementation, the sound pickup device is disposed on the merchandise display rack, and may be a directional microphone for receiving the sound directly in front. In this way, you can receive sounds related to the product, such as: "This is beautiful," "Do not buy this," etc., and identify people based on the voiceprint of the voice, and perform semantic analysis on the sound to generate Shopping preferences, for example, when the voice is "this is beautiful," the semantic analysis result is positive, so the corresponding product is recorded in shopping preferences; when the voice is "don't buy this", the semantic analysis result is negative evaluation. , so the corresponding product records are removed from the shopping preferences, and even the weight value can be used to record each product in the shopping preferences, and each product corresponds to a weight value, when the semantic analysis result is a positive evaluation, Increase the weight value corresponding to the commodity. Conversely, when the semantic analysis result is negative evaluation, reduce the weight value corresponding to the commodity. Finally, the weight value recorded in the shopping preferences can be used as a basis for judging people's preference for merchandise.

The processing module 150 is configured to embed the shopping preferences of the corresponding voiceprint features into the customer message having the same voiceprint feature, and embed the shopping path and the stay time generated according to the sensed shoe print feature and the weight data. In the customer message of the same shoe print feature and weight data. Simply put, the shopping preferences, shopping path and stay time are embedded in the corresponding customer message. In this way, the customer's movements in the store can be analyzed to provide the merchant with the basis for purchasing or providing customized services. In particular, when there are more than one customer message with the same shoe print feature and weight data, the processing module 150 may match the facial features and voiceprint features, or at least one of them to determine the shopping path. And which customer message is embedded in the stay time.

In particular, the customer database 110, the sensing module 120, the path module 130, the voice recognition module 140, and the processing module 150 included in the system of the present invention can be implemented in various ways. The implementation includes software, hardware, or any combination thereof. For example, in some embodiments, each module may be implemented by using one or more of software and hardware, and in addition, the system may be partially or completely Based on hardware, for example, one or more modules in the system can pass through integrated circuit chips, system on chip (SoC), Complex Programmable Logic Device (CPLD), and on-site. Field Programmable Gate Array (FPGA) and so on.

The invention can be a system, method and/or computer program. The computer program can include a computer readable storage medium having computer readable program instructions for causing a processor to implement various aspects of the present invention. The computer readable storage medium can be tangible for retaining and storing instructions for use by the instruction execution device. device. The computer readable storage medium can be, but is not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples (non-exhaustive lists) of computer readable storage media include: hard disks, random access memory, read only memory, flash memory, optical disks, floppy disks, and any suitable combination of the foregoing. A computer readable storage medium as used herein is not to be interpreted as a transient signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (eg, an optical signal through a fiber optic cable), or through a wire The electrical signal transmitted. In addition, the computer readable program instructions described herein can be downloaded from a computer readable storage medium to various computing/processing devices or downloaded over a network such as the Internet, a local area network, a wide area network, and/or a wireless network. Go to an external computer device or an external storage device. The network may include copper transmission cables, fiber optic transmissions, wireless transmissions, routers, firewalls, switches, hubs, and/or gateways. A network card or network interface in each computing/processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage on a computer readable storage medium in each computing/processing device in. Computer program instructions for performing the operations of the present invention may be combined language instructions, instruction set architecture instructions, machine instructions, machine related instructions, microinstructions, firmware instructions, or source code or destination code written in any combination of one or more programming languages. (Object Code), the programming language includes object-oriented programming languages such as: Common Lisp, Python, C++, Objective-C, Smalltalk, Delphi, Java, Swift, C#, Perl, Ruby, PHP, etc., as well as conventional Procedural programming language, such as C language or similar programming language. The computer readable program instructions can be executed entirely on the computer, partly on the computer, as a standalone software, partly on the client computer, partly on the remote computer, or entirely on the remote computer or server. Execute on.

Next, please refer to "Fig. 2", which is a flow chart of a method for customer dynamic tracking according to the present invention. The steps include: providing an inductive floor, a sound pickup device and an entrance gate, wherein the entrance gate includes an entrance sensor ( Step 210); the entrance sensor senses and records facial features, voiceprint features, weight data, and shoe print features when the person enters the entrance gate to store to the customer database as a customer message corresponding to the person (steps) 220); the sensing floor continuously senses the shoe print feature and the weight data, and continuously generates a corresponding shopping path and dwell time according to the sensed shoe print feature and the weight data (step 230); receiving the corresponding voice message through the sound receiving device, And performing voiceprint and semantic analysis on the voice message to obtain shopping preferences corresponding to the voiceprint feature (step 240); embedding the shopping preferences corresponding to the voiceprint feature into the customer message having the same voiceprint feature, and based on the sensed Shoe print features and weight data generated by the shopping path and residence time, embedded with the same shoe print features and weight Customers message data (step 250). Through the above steps, the facial features, the voiceprint features, the weight data and the shoe print features can be used as customer information, and the corresponding shopping path and stay time can be generated according to the shoe print characteristics and weight data sensed by the sensing floor. And receiving voice messages for voiceprint and semantic analysis to obtain corresponding shopping preferences, so as to embed the shopping preferences, shopping paths and staying time into the customer information, and provide the merchant as the basis for the purchase and sales strategy adjustment to achieve the unused wireless Under the premise of RFID technology, track customer dynamics and learn about shopping preferences.

The following description will be made by way of example with reference to "Fig. 3" and "Fig. 4". Please refer to "Fig. 3" and "Fig. 3" for a schematic view of a store to which the present invention is applied. In practical implementation, the entrance of the store to which the present invention is applied is provided with an entrance gate 300, which is provided with a plurality of different inlet sensors, such as: CCD, sound sensing, weight sensing, shoe print sensing And so on, and a plurality of sensing floors 310 are disposed in the store, and the sensing floor 310 can also be disposed at the entrance as one of the entrance sensors for sensing the shoe print of the customer (also referred to as a person). feature. In addition, a merchandise display rack 320 is disposed in the store, and each merchandise display rack 320 includes a sound pickup device, an image pickup device, and the like. The sound receiving device can be a directional microphone for receiving the sound in front of the voice, so as to receive the voice message for voiceprint and semantic analysis, and then as a basis for determining the shopping preference, for example, analyzing the person by voiceprint analysis, analyzing and judging the person by semantic analysis. A positive or negative evaluation of the item on the merchandise display rack 320 is made. Next, the camera device is configured to capture the pupil position of the person to analyze whether the person looks at the product and generates a gaze time, so as to adjust the shopping preference according to at least one of the gaze time and the stay time of the product display rack 320, for example, when the camera device When the person looks at the product and keeps watching for one minute or longer, it can be determined that the product of the product display rack 320 is favorite, so that the product of the product display rack 320 is added to the shopping preference.

As shown in "Fig. 4", "Fig. 4" is a schematic diagram of generating a shopping path by applying the present invention. When the person enters the store, the sensor floor 310 that is stepped on sequentially transmits the sensed shoe print feature and weight data to the path module 130. At this time, the path module 130 can continuously generate a shopping path and a stay time based on the sensed shoe print characteristics and weight data. In practical implementation, each of the sensing floor 310 and the merchandise display racks (411-414) may have corresponding coordinates to know their positions, so that it can be known that the sensing floors 310 are adjacent to each other, and the merchandise display racks (411~) 414) What are the sensing floors 310 in front and so on. Taking "Fig. 4" as an example, the inductive floor coordinates in the upper left corner can be "(1,1)". This coordinate is a two-dimensional coordinate. The former represents the horizontal axis and the latter represents the vertical axis. Therefore, the sensing floor 310 is "(1,1)" to "(1,6)" and "(2,1)" to "(2,6)" from left to right and from top to bottom. And so on to "(7,6)". The merchandise display rack 411 has three coordinates "(1, 2), (2, 2) and (3, 2)", and the merchandise display rack 412 has three coordinates "(1, 6), (2, 6) respectively. And (3,6)", the merchandise display rack 413 has three coordinates "(5, 2), (6, 2) and (7, 2)" respectively, and the product display rack 414 has three coordinates respectively ((5, 6), (6, 6) and (7, 6). By the coordinates described above, it can be determined that the sensing floor 310 in front of the product display rack 414 is "(5, 5), (6, 5) and (7, 5)", respectively. At this time, when the person stays on the sensor floor 310 with the coordinates "(6, 5)" for 2 minutes, it means that the stay is 2 minutes before the product display rack 414, so the 2 minutes can be used as the stay time.

Next, suppose that the personnel step on each of the sensing floors (such as "Fig. 4" to indicate the inductive floor 421 that has been stepped on by the dots). The path module 130 can generate a corresponding shopping path based on the coordinates of the sensing floors 421 and record the dwell time at each coordinate. It is important to note that the same shopping path must have the same shoe print characteristics and weight data. In this way, the customer dynamics can be known by the shopping path without using the radio frequency identification technology, and the stay time is used as a basis for evaluating the customer's shopping preferences. For example, the longer the stay time, the higher the preference. In addition, in addition to evaluating the shopping preferences by the dwell time, the processing module 150 also embeds the shopping preferences obtained by the voiceprint and semantic analysis into the customer information, and also serves as a basis for evaluating the customer's shopping preferences.

In summary, it can be seen that the difference between the present invention and the prior art is that by sensing facial features, voiceprint features, weight data, and shoe print features as customer messages, and continuing to sense the footprint characteristics and weight based on the sensing floor. The data generates corresponding shopping paths and residence times, and receives voice messages for voiceprint and semantic analysis to obtain corresponding shopping preferences, so as to embed shopping preferences, shopping paths and staying time into customer messages, and provide merchants as purchase and sales strategy adjustments. The basis for achieving the tracking of customer dynamics and knowing the shopping preferences without using the radio frequency identification technology, by which a technical means can solve the problems of the prior art, thereby achieving the unused radio frequency identification technology. Under the premise of tracking customer dynamics and knowing the technical benefits of shopping preferences.

While the present invention has been described above in the foregoing embodiments, it is not intended to limit the invention, and the invention may be modified and modified without departing from the spirit and scope of the invention. The scope of patent protection shall be subject to the definition of the scope of the patent application attached to this specification.

110‧‧‧Customer Database

120‧‧‧Sensor module

130‧‧‧Path Module

140‧‧‧Voice recognition module

150‧‧‧Processing module

300‧‧‧ entrance gate

310‧‧‧Induction floor

320‧‧‧ merchandise display rack

411~414‧‧‧Commodity display rack

420‧‧‧ shopping path

421‧‧‧The sensor floor that has been stepped on

Step 210‧‧‧ providing a plurality of sensing floors, at least one sounding device and an inlet gate, wherein the inlet gate comprises a plurality of inlet sensors

Step 220‧‧‧ The entrance sensor senses and records a facial feature, a voiceprint feature, a weight data, and a shoe print feature to be stored in a customer database as a person enters the entrance gate a customer message corresponding to the person

Step 230‧‧‧ The sensing floor continuously senses the shoe print feature and the weight data, and continuously generates a corresponding shopping path and a dwell time according to the sensed shoe print feature and the weight data

Step 240 ‧ ‧ receiving a corresponding voice message through the sound receiving device, and performing voiceprint and semantic analysis on the voice message to obtain a shopping preference corresponding to the voiceprint feature

Step 250‧‧‧ inserting the shopping preference corresponding to the voiceprint feature into the customer message having the same voiceprint feature, and the shopping path generated based on the sensed shoeprint feature and the weight data The dwell time is embedded in the customer message having the same shoe print feature and the weight data

Figure 1 is a block diagram of the system of the customer dynamic tracking system of the present invention. 2 is a flow chart of a method for the customer dynamic tracking method of the present invention. Figure 3 is a schematic illustration of a store to which the present invention is applied. Figure 4 is a schematic diagram of generating a shopping path by applying the present invention.

Claims (10)

A customer dynamic tracking system, comprising: a customer database for storing at least one customer message; a sensing module, the sensing module comprising a plurality of inlet sensors and disposed at an inlet gate for When a person enters the entrance gate, sensing and recording a facial feature, a voiceprint feature, a weight data, and a shoe print feature for storage to the customer database as the customer message corresponding to the person; a module for electrically connecting a plurality of sensing floors, each sensing floor is configured to sense the shoe printing feature and the weight data, and continuously generate a corresponding shopping path according to the sensed shoe printing feature and the weight data a pause time; a voice recognition module electrically connected to at least one sound pickup device for receiving a corresponding voice message through each of the sound pickup devices, and performing voiceprint and semantic analysis on the voice message to obtain a corresponding voiceprint feature a shopping preference; and a processing module for embedding the shopping preference corresponding to the voiceprint feature in the customer message having the same voiceprint feature And cart path generated based on the sensed characteristic of the shoe prints and the weight data and the residence time, together with the same message embedded in said customer and wherein the weight of the shoe print data. According to the customer dynamic tracking system of claim 1, wherein the sound collecting device is disposed on at least one product display rack, and each of the product display racks further includes a camera device, and the processing module is electrically connected to the camera device for A person's pupil position is retrieved to analyze whether the person is looking at the item and generating a gaze time. According to the customer dynamic tracking system of claim 2, the processing module adjusts the shopping preference according to at least one of the gaze time and the stay time of the merchandise display rack. The customer dynamic tracking system according to claim 1, wherein the sound receiving device is a directional microphone for receiving sound in front of the sound. According to the customer dynamic tracking system of claim 1, wherein the sensing floor comprises a silicone layer, an acryl layer, a photosensitive coupling element, and at least one infrared light emitting diode disposed on a side of the sensing floor. And, when the silicone layer is pressed and contacts the acryl layer, the infrared rays propagating in the acryl layer are scattered, and the scattered infrared rays are extracted by the photosensitive coupling element to generate the shoe print feature. A customer dynamic tracking method, the method comprising: providing a plurality of sensing floors, at least one sounding device, and an inlet gate, wherein the inlet gate comprises a plurality of inlet sensors; the entrance sensor enters the entrance gate by a person Sensing and recording a facial feature, a voice feature, a weight data, and a shoe print feature for storage in a customer database as a customer message corresponding to the person; the sensing floor continuously sensing the shoe print feature And the weight data, and continuously generating a corresponding shopping path and a dwell time according to the sensed shoe print feature and the weight data; receiving a corresponding voice message through the sound receiving device, and performing sound recording on the voice message And semantic analysis to obtain a shopping preference corresponding to the voiceprint feature; and embedding the shopping preference corresponding to the voiceprint feature in the customer message having the same voiceprint feature, and based on the sensed shoeprint feature And the shopping path and the dwell time generated by the weight data are embedded together with the same shoe print feature and the weight number The customer message. According to the customer dynamic tracking method of claim 6, wherein the sound collecting device is disposed on at least one merchandise display rack, and each merchandise display rack further includes a camera device for capturing a pupil position of the person to analyze the Whether the person looks at the product and generates a gaze time. The customer dynamic tracking method according to claim 7, wherein the method further comprises the step of adjusting the shopping preference according to at least one of the gaze time of the merchandise display rack and the dwell time. According to the customer dynamic tracking method of claim 6, wherein the sound receiving device is a directional microphone for receiving the sound in front. According to the customer dynamic tracking method of claim 6, wherein the sensing floor comprises a silicone layer, an acryl layer, a photosensitive coupling element, and at least one infrared light emitting diode disposed on a side of the sensing floor. And, when the silicone layer is pressed and contacts the acryl layer, the infrared rays propagating in the acryl layer are scattered, and the scattered infrared rays are extracted by the photosensitive coupling element to generate the shoe print feature.