Method and system for improving scenic spot experience of tourists in tourist area, medium and processor

Abstract

The invention provides a method for improving scenic spot experience of tourists in a tourist area, and the method comprises the steps: S1, automatically collecting images of the tourists, analyzing the images of the tourists, and obtaining the emotional states of the tourists; s2, separating and obtaining a first picture of the tourist in the first emotional state; s3, obtaining the time and geographic position corresponding to the first photo; s4, associating the geographic position shot by the first photo with a related geographic position on a map, and marking on the map; s5, sequentially arranging the marked related geographic positions on the map according to the sequence of the shooting time of all the first photos, and generating a card punching track album; the card punching track album is generated by automatically collecting the images of the tourists, the shot photos are closer to the real moods of the tourists, the moments worthy of being minded are not missed, the recall experience feeling of the tourists is higher through the generated album, the tourists do not need to take photos into account, the tourists can play in a whole body mode, and the tourism experience depth and level are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of tourism check - in, and particularly to a method, system, medium and processor for improving the tourist attraction experience in a tourist area. Background Art

[0002] Although tourist areas continuously update their scenic spot ideas to increase their popularity, from the later investigation and feedback, although the popularity of the tourist area has increased, there is still a gap between the depth of experience of most tourists and the expected ideal level. That is, the experience of most tourists is limited to the level of joining in the fun.

[0003] In view of this, a method, system, medium and processor for improving the tourist attraction experience in a tourist area are needed. Summary of the Invention

[0006] Aiming at the problem that the depth of experience of most tourists in the prior art is insufficient, that is, the experience of tourists is mostly limited to the level of joining in the fun, the present invention provides a method, system, medium and processor for improving the tourist attraction experience in a tourist area, which can enhance the depth and level of the tourism experience. The specific technical solutions are as follows:

[0007] A method for improving the tourist attraction experience in a tourist area includes:

[0008] S1: Automatically collect the images of tourists, and analyze the images of tourists to obtain the emotional state of tourists;

[0009] S2: Separate and obtain the first photos of tourists in the first emotional state;

[0010] S3: Obtain the time and geographical location corresponding to the first photos;

[0011] S4: Associate the geographical location where the first photos are taken with the relevant geographical locations on the map and mark them on the map;

[0012] S5: Arrange the marked relevant geographical locations in chronological order of the shooting times of all the first photos on the map to generate a check - in track photo album;

[0013] By automatically collecting images of tourists to generate a check - in track photo album, the photos taken are closer to the real mood of tourists and will not miss the memorable moments; the photo album generated by screening the photos with the best emotional state makes the tourists' memory experience stronger; it allows tourists to focus on playing without worrying about taking pictures, enhancing the depth and level of the tourism experience.

[0014] Further, when analyzing the images of tourists, only part of the video data is collected for emotional analysis; the part of the images collected adopts a timed collection method.

[0015] Further, separating and obtaining the first photo of the tourist in the first emotional state includes the following steps:

[0016] Separating and obtaining a number of second photos of the tourist in the first emotional state;

[0017] Finding the photo with the best state among the second photos as the third photo;

[0018] Taking the time corresponding to the third photo as the acquisition benchmark, re-acquiring images from the video data within the first and third times before and after to obtain a number of fourth photos;

[0019] Comparing and analyzing the number of fourth photos and the third photo together, and obtaining the photo with the best state as the first photo.

[0020] Further, obtaining the photo with the best state means calculating the comprehensive score of the photos, and taking the photo with the highest score as the photo with the best state.

[0021] Further, the comprehensive score calculation includes the following steps:

[0022] Data standardization, performing standardization processing on the original data x of each index ij to obtain y ij , to eliminate the influence of dimension, where i represents the photo serial number and j represents the index serial number;

[0023] Calculating information entropy, the information entropy calculation formula for the jth index is as follows:

[0024]

[0025] where m is the number of photos;

[0026] Calculating entropy weight, the entropy weight calculation formula for the jth index is as follows:

[0027] n is the number of indexes;

[0028] Calculating the comprehensive score, the formula is as follows:

[0029]

[0030] where x i is the score of each index after standardization; n i is the number of indexes of the ith photo; represents the total number of indexes of all photos participating in the comparison this time.

[0031] Further, it also includes the following steps:

[0032] S6: Analyze each of the first photos in the check-in track photo album to obtain the natural environmental conditions of each first photo; obtain the corresponding poems and songs according to the natural environmental conditions and display them in association with the first photos, which can help tourists with basic photo album editing;

[0033] S7: Display the check-in track photo album to tourists and obtain the photo selection information input by tourists, and regenerate the check-in track photo album according to the selection information.

[0034] Furthermore, it also includes the following steps:

[0035] S8: Automatically collect the images of tourists and analyze the images of tourists to obtain the body movement information of tourists; if tourists make specific body movements, take pictures of tourists to obtain the fifth photo;

[0036] S9: Obtain the time, geographical location and corresponding poems and songs corresponding to the fifth photo, and display them in association in the check-in track photo album in the order of the shooting time.

[0037] A system for improving the tourist scenic spot experience in a tourist area, which is applied to the method for improving the tourist scenic spot experience in the tourist area described above, includes:

[0038] The first collection module is used to automatically collect the images of tourists and analyze the images of tourists to obtain the emotional state of tourists;

[0039] The first separation module is used to separate and obtain the first photos of tourists in the first emotional state;

[0040] The second collection module is used to obtain the time and geographical location corresponding to the first photo;

[0041] The first annotation module is used to associate the geographical location where the first photo is taken with the relevant geographical location on the map and annotate it on the map;

[0042] The generation module is used to arrange the marked relevant geographical locations in sequence on the map according to the shooting time sequence of all the first photos to generate a check-in track photo album;

[0043] By automatically collecting the images of tourists to generate a check-in track photo album, the photos taken are closer to the real mood of tourists, and the memorable moments will not be missed. The generated photo album gives tourists a stronger sense of memory experience. At the same time, tourists don't need to worry about taking pictures and can fully immerse themselves in playing, enhancing the depth and level of the tourism experience.

[0044] A computer-readable storage medium, the computer-readable storage medium including a stored program, wherein, when the program runs, it controls the device where the computer-readable storage medium is located to execute the method for improving the tourist attraction experience in the tourist area described above.

[0045] A processor, the processor being used to run a program, wherein, when the program runs, it executes the method for improving the tourist attraction experience in the tourist area described above.

[0046] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0047] 1. In this application, the camera in the tourist area is used to automatically collect images of tourists. After analyzing the expressions of the tourists in the images, instant photos in a happy or excited or other state worth taking pictures of are obtained, and a check-in track photo album is automatically generated based on this. Compared with manual photography and editing methods such as mobile phones and cameras, many accidental happy moments of tourists can be collected, and each happy moment of tourists can be recorded from an objective perspective. Compared with the posed photos taken by cameras or mobile phones, the photos taken by the camera are closer to the true mood of tourists, and the generated photo album gives tourists a stronger sense of memory experience and does not miss any happy moment. At the same time, the solution of this application allows tourists not to constantly worry about where it is appropriate to take pictures and only focus on check-in, enabling tourists to fully immerse themselves in playing in the tourist area without having to worry about taking pictures, thus enhancing the depth and level of the tourism experience. Description of the Drawings

[0048] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the description of the specific embodiments or the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts do not necessarily draw according to the actual scale.

[0049] Figure 1 It is a schematic flowchart of a method for improving the tourist attraction experience in the tourist area;

[0050] Figure 2 It is a schematic structural diagram of a system for improving the tourist attraction experience in the tourist area. Detailed Embodiments

[0051] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0052] It should be understood that when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0053] It should also be understood that the terms used in the specification of the present invention are merely for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an", and "the" are intended to include the plural forms.

[0054] It should be further understood that the term "and / or" used in the specification of the present invention and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0055] Embodiment 1

[0056] As Figure 1 shown is a schematic flowchart of a method for improving the tourist attraction experience in a tourist area, including the following steps:

[0057] S1: Automatically collect the images of tourists and analyze the images of tourists to obtain the emotional states of tourists. Specifically, it includes the following steps:

[0058] 1. Image acquisition

[0059] Hardware deployment

[0060] Camera installation: Reasonably install cameras in areas where tourists may appear, such as at the scenic spot entrance and popular attractions. The cameras need to have a high-definition resolution (such as 1080p and above) to ensure that the collected images can clearly display the facial features and full-body limb movements of tourists. For example, install multiple cameras in the queuing areas of various amusement facilities in a large theme park to ensure full coverage.

[0061] Network connection: Connect the cameras to the data processing center through a wired network (such as Ethernet) or a wireless network (such as Wi-Fi, 4G / 5G) to transmit the collected image data in real time.

[0062] Video acquisition software settings. Generally, the acquisition frequency is as follows, and the collected video data is transmitted to the data processing center for storage.

[0063] 24fps (frames per second): The frame interval is about 41.67 milliseconds (1000 milliseconds ÷ 24 ≈ 41.67 milliseconds). This is a common frame rate in the film industry, which can provide a natural, cinematic visual experience because it is close to the human visual system's comfortable range for dynamic images, suitable for telling stories, and simulating the rhythm of human eyes observing motion in real life.

[0064] 25fps: The frame interval is 40 milliseconds (1000 milliseconds ÷ 25 = 40 milliseconds). It is widely used in PAL television systems. Television broadcasts in Europe, China and other regions often use this frame rate. It can better adapt to the power grid frequency (50Hz), reduce flickering, and ensure video playback stability.

[0065] 30fps: The frame interval is about 33.33 milliseconds (1000 milliseconds ÷ 30 ≈ 33.33 milliseconds). This is a common frame rate in the fields of online video and game live broadcasting. It strikes a balance between fluency and data volume, providing relatively smooth visual effects without generating excessive data volume, which is convenient for network transmission and device processing.

[0066] 60fps: frame interval is 16.67 milliseconds (1000 milliseconds ÷ 60 ≈ 16.67 milliseconds). High frame rate can significantly improve the smoothness of the picture, especially in fast-action scenes, such as live sports events and high-dynamic game scenes, allowing the audience to clearly capture the details of fast movements. It is often used in scenes that pursue the ultimate smooth experience.

[0067] 120fps and above: The frame interval is shorter, such as 8.33 milliseconds for 120fps (1000 milliseconds ÷ 120 ≈ 8.33 milliseconds). This ultra-high frame rate is used in professional film and television production, high-end games, and some special scenes that need to capture extremely fast movements, such as high-speed moving object analysis and virtual reality content creation. It provides extremely smooth and realistic visual experience, but has extremely high requirements for shooting equipment, storage, and transmission.

[0068] Video collection trigger: By setting up sensors (such as infrared sensors), when a tourist is detected entering a specific area, the camera is triggered to immediately collect the tourist video. For example, an infrared sensor is set at the entrance of a small exhibition hall in a scenic area, and video images are collected immediately when a tourist enters.

[0069] Furthermore, when analyzing the images of tourists, only part of the images are collected from the video data for emotion analysis.

[0070] Furthermore, the acquisition of some images can be done in a timed acquisition mode: a frame of image can be acquired at a certain interval (such as 1-5 seconds) for emotion analysis. This can obtain sufficient image data without generating too much redundant information and wasting computing resources, thus speeding up the analysis and calculation.

[0071] Further, if it is detected that the speed of the tourist's facial expression change accelerates, the image acquisition frequency usually needs to be increased when collecting partial images. For example, it is changed to collect one frame every 0.1 second.

[0072] 2. Image preprocessing

[0073] The collected original images may have various problems and need to be preprocessed to improve the accuracy of subsequent emotion analysis.

[0074] Gray conversion: Convert the color image into a grayscale image to simplify the image data and reduce the amount of calculation. Most emotion analysis algorithms are based on the features of grayscale images. For example, to convert an RGB format image into a grayscale image, the formula is: Gray = 0.299*R + 0.587*G + 0.114*B, where R (Red) represents the value of the red channel and represents the intensity of the red component in the image. In the common RGB color model, the color of each pixel is composed of three color components: red, green, and blue. The R value reflects the content of the red component of this pixel, and its value range is usually 0 - 255. 0 means no red, and 255 means the strongest intensity of red. G (Green) represents the value of the green channel and reflects the intensity of the green component in the image. Green often carries rich detailed information in natural scenes and various images. Similarly, the G value range is 0 - 255, and the numerical size determines the intensity of green. B (Blue) refers to the value of the blue channel and is used to represent the intensity of the blue component in the image. Similar to R and G, the B value varies between 0 and 255 and determines the intensity of blue in the pixel. The human eye has different sensitivities to different colors, being most sensitive to green, followed by red, and relatively insensitive to blue. By assigning different weights (0.299, 0.587, 0.114) to R, G, and B, this formula can better conform to the human eye's perception of grayscale and retain the visual brightness information as much as possible when converting the color image into a grayscale image.

[0075] Image denoising: Use methods such as Gaussian filtering and median filtering to remove the noise in the image, such as salt-and-pepper noise, Gaussian noise, etc., to make the image smoother. For example, use a Gaussian filter to perform a convolution operation on the image to eliminate the noise generated by factors such as the camera's photosensitive element.

[0076] Face detection and alignment: Use face detection algorithms (such as Haar cascade detector, MTCNN based on deep learning, etc.) to locate the position of the face in the image and align the face image to unify the pose and size of the face. Usually, the face image is normalized to a fixed size (such as 112×112 pixels) for subsequent feature extraction.

[0077] 3. Selection and training of emotion analysis model

[0078] Select a pre-trained model: Currently, there are many mature pre-trained sentiment analysis models available for selection, such as convolutional neural network (CNN) models like VGG-16 and ResNet trained on the FER2013 dataset. These models have been trained with a large amount of data on public datasets and have a certain ability to recognize emotions.

[0079] Fine-tune the model: If the performance of the pre-trained model is not satisfactory in a specific scenario, the model can be fine-tuned using the tourist image data collected by yourself. First, collect and annotate a large amount of tourist emotion image data, and annotate the emotion categories (such as happy, sad, angry, surprised, etc.). Then, adjust the last few layers (such as the fully connected layer) of the pre-trained model to adapt to the new classification task. Use the annotated tourist image data to train the model and adjust the model parameters to make it have better performance in the tourist emotion analysis task.

[0080] Of course, other existing models and datasets can also be used for training, as long as they can meet the emotion analysis of face images.

[0081] 4. Implementation of Emotion Analysis

[0082] Feature extraction: Input the preprocessed face image into the trained emotion analysis model, and the model will automatically extract the features in the image, such as facial muscle texture, expression changes and other feature information.

[0083] Emotion classification: According to the extracted features, the model predicts the emotion state of tourists through a classification algorithm (such as the Softmax classifier) and outputs the probability values belonging to each emotion category. For example, the output result may be: happy (0.8), sad (0.1), angry (0.05), surprised (0.05), indicating that the probability of this tourist being in a happy mood is 80%.

[0084] 5. Result Display and Application

[0085] Real-time display: On the monitoring screen of the scenic area management center, the collected tourist images and their corresponding emotion analysis results are displayed in real time. For example, tourists with different emotions are marked with different colored borders, red for angry, green for happy, etc., to facilitate the management staff to timely understand the tourist emotion state.

[0086] Data analysis and decision-making: Statistically analyze the tourist emotion data over a period of time to understand the emotion distribution of tourists at different times and different scenic spots. The scenic area can optimize the scenic spot settings, service quality, etc. according to these data. For example, add rest facilities or optimize the guided tour service at scenic spots where tourists are in a bad mood.

[0087] S2: Separate and obtain the first photo of the tourist in the first emotional state.

[0088] Furthermore, the first emotional state is set to different emotional states according to different scenic spots in the tourist area; that is, at the first scenic spot, the first emotion includes excitement, awe and pride; at the second scenic spot, the first emotion changes to shock and surprise; at the third scenic spot, the first emotion changes to joy, peace and happiness, etc.; by matching the first emotion to different emotional states at different scenic spots, the amount of calculation can be reduced and the speed of separating and obtaining the first photo can be accelerated. The specific examples include the following categories:

[0089] 1. Visit famous attractions

[0090] Emotions: Full of excitement, awe and pride. Excited to finally be at a place you have longed to visit, in awe of the history and culture or spectacular sights it carries, and proud to be able to visit it in person.

[0091] Facial expressions: The eyes are usually wide open, with the gaze focused on the scenic spots, revealing curiosity and amazement; the corners of the mouth are raised, revealing a happy smile, and the mouth may be slightly open, showing uncontrollable excitement.

[0092] 2. Surround yourself with rare natural landscapes

[0093] Emotion: More of shock and surprise. Shocked by the wonders of nature, and overjoyed to witness such a rare sight.

[0094] Facial expressions: Most of the expressions are surprise and intoxication, with eyes wide open and mouth unconsciously open to express inner shock; facial muscles relaxed, revealing an intoxicated look.

[0095] 3.Important festivals:

[0096] Mood: Full of joy, peace and happiness. The festive atmosphere and reunion with family bring warm and pleasant feelings.

[0097] Facial expressions: They are all smiles, with cheeks slightly flushed with happiness, and their eyes sparkling with joy. They may also interact closely with people around them, with expressions full of warmth.

[0098] 4. Important moments in life:

[0099] Graduation ceremony: mixed emotions, with the joy and pride of completing studies, and the reluctance to leave classmates and teachers. In terms of facial expressions, the smile is confident and bright, showing the vision of the future; but sometimes there is a hint of sadness, and tears may flash in the eyes.

[0100] 5. Wedding ceremony: full of happiness and sweetness, the expectation for a new life reaches its peak. The newlyweds always have a happy smile on their faces, their eyes are full of affection, and when they gaze at each other, they reveal endless love. The guests around them will also smile with blessings.

[0101] 6. Birthday Party: Mainly with lively and satisfied emotions. The birthday star has a big happy laugh on the face, enjoying everyone's blessings, while others have friendly and happy smiles, together creating a lively atmosphere.

[0102] 7. Trying New Activities:

[0103] Emotions: Excited, nervous and full of a sense of achievement. Excited about challenging new things, nervous due to the uncertainty of the unknown, and having a strong sense of achievement after a successful attempt.

[0104] Facial Expressions: Before the attempt, may slightly frown and have tightly closed lips, showing a bit of nervousness; during the attempt, the eyes are focused and the expression is serious; after success, will show a bright smile, with pride and excitement in the eyes.

[0105] 8. Tasting Specialties:

[0106] Emotions: Pleased and satisfied. The gustatory enjoyment brought by the delicious food makes people in a happy mood.

[0107] Facial Expressions: When tasting, may slightly close the eyes, showing an enjoyable expression, savoring carefully; then open the eyes, with the corners of the mouth turned up, showing a satisfied smile, and may even make an exclamation sound.

[0108] 9. Gathering with Relatives and Friends:

[0109] Emotions: Excited, affectionate and warm. The joy of reunion or gathering, as well as the deep emotional connection makes people's hearts full of warmth.

[0110] Facial Expressions: Have a warm smile on the face, with affection and care in the eyes, may hug each other, pat on the shoulder, communicate vividly with expressions, and the whole face emits a happy glow.

[0111] 10. Being Inspired by Art:

[0112] Emotions: Immersed in being touched, resonating or inspired. Being touched by the connotation of the art work, generating strong emotional reactions.

[0113] Facial Expressions: Look focused and engaged, may slightly frown or have a blank look in the eyes, immersed in thinking; may also have glistening touched tears in the eyes, with the corners of the mouth slightly trembling, showing inner waves.

[0114] 11. Displaying Personal Elegance:

[0115] Emotions: Confident and proud. Being confident in one's dressing and state, hoping to be recognized by others.

[0116] Facial Expressions: Smile confidently, with firm and bright eyes, may pose in various elegant or fashionable postures, with the head slightly raised, showing the best state.

[0117] 12. Record life insights:

[0118] Emotion: calm, thinking or feeling. When having insights into life, the heart is relatively calm and the thoughts are immersed in thinking.

[0119] Facial expression: The expression is relatively peaceful, with a thoughtful look in the eyes. The corners of the mouth may turn up slightly or be pursed, showing unique thinking about life.

[0120] Furthermore, the step of separating and obtaining the first photo of the tourist in the first emotional state includes the following steps:

[0121] Separate and obtain a number of second photos of the tourist in the first emotional state;

[0122] Find the one with the best state among the second photos as the third photo;

[0123] Taking the time corresponding to the third photo as the acquisition benchmark, re-acquire images from the video data within the first and third time periods before and after to obtain a number of fourth photos;

[0124] Compare and analyze the number of fourth photos and the third photo together to obtain the photo with the best state as the first photo.

[0125] Furthermore, the third time is the time interval for collecting some images in step S1.

[0126] For example: Suppose there are 10 second photos of the tourist in an excited state obtained at the first scenic spot (the time interval for collecting some images in step S1 is 1 frame per second, and the tourist is continuously in an excited state for 10 seconds; or the tourist is in an excited state in several time periods at the first scenic spot, and there are a total of 10 second photos in several time periods); Compare the 10 second photos with each other to find the one with the best state as the third photo, and the shooting time point of the third photo is 10:15 am, and the video acquisition frequency is 24 fps (frames per second); Taking 10:15 as the acquisition benchmark, re-acquire the pictures included in the video data between 10:14 and 10:16 to obtain 48 fourth photos; Compare and analyze the 49 fourth photos and the third photo together. If it is found that the photo at 10:15:30 is the best, then take this photo as the first photo.

[0127] Furthermore, the step of obtaining the photo with the best state includes the following steps:

[0128] 1. Facial expression analysis

[0129] Quantification of the degree of smile:

[0130] Key point detection: Utilize the dnn module of OpenCV or a dedicated facial key point detection library (such as dlib) to locate the key feature points on the face, such as the corners of the mouth, around the eyes, etc. For example, the dlib library can accurately locate 68 facial key points. By analyzing the angle and distance of the upward curve of the corners of the mouth, the degree of the smile can be quantified. The larger the upward angle of the corners of the mouth and the wider the distance between the corners of the mouth, the brighter the smile usually is.

[0131] Muscle movement analysis: The smile on the face involves the movement of muscles such as the zygomatic major muscle. Some advanced computer vision techniques can more accurately evaluate the intensity of the smile by analyzing the skin texture changes caused by these muscles. For example, use the optical flow method to track the tiny displacements of the skin during facial muscle movement, thereby more precisely measuring the degree of the smile.

[0132] Evaluation of the vitality of the eyes: When a person is happy, their eyes are usually bright and energetic.

[0133] Pupil state: Detect the size and brightness of the pupils. In a happy mood, the pupils may be relatively dilated, and the light reflected by the eyes is brighter. By analyzing the proportion of the pupil area in the image and the average brightness of the pupil region, the vitality of the eyes is evaluated.

[0134] Eye muscle movement: Detect the movement of the muscles around the eyes (such as the orbicularis oculi muscle). When a person smiles happily, the orbicularis oculi muscle contracts, causing wrinkles (crow's feet) to appear around the eyes and the eyes to squint. By analyzing the characteristics generated by these eye muscle movements, such as the depth of the wrinkles and the degree of eye squinting, the degree of happiness conveyed in the eyes is judged.

[0135] 2. Composition and picture aesthetic feeling

[0136] Subject position and proportion:

[0137] Golden ratio rule: Judge whether the position of the tourist in the photo conforms to the golden ratio. Divide the photo image according to the golden ratio (about 1:0.618). Ideally, the key parts of the tourist's face or body (such as the eyes) are near the golden ratio points. Such a composition is more aesthetically pleasing and visually attractive. It can be evaluated whether it conforms to the golden ratio by calculating the distance relationship between the key points on the tourist's face and the photo boundary.

[0138] Picture balance: Analyze whether the distribution of the tourist and the surrounding environmental elements in the photo is balanced. Avoid the situation where the tourist is on one side, while there is a large amount of blank space or elements are overly concentrated on the other side. Judge the balance of the picture by calculating the pixel density of different areas of the picture or the center of gravity of the object distribution.

[0139] Background simplicity and relevance:

[0140] Background complexity: Metrics such as image entropy are used to measure the complexity of the background. The higher the image entropy, the more details and variations in the background, which may distract from the main subject (the tourist). A lower image entropy means the background is relatively simple and can better highlight the main subject.

[0141] Relevance between background and theme: Using image classification or object detection techniques, identify the objects in the background and judge their relevance to the theme of tourists being happy. For example, in a photo taken at a tourist attraction, the presence of the iconic building of the attraction in the background will enhance the theme expression of the photo; while irrelevant clutter may reduce the photo quality.

[0142] 3. Light and color

[0143] Light uniformity and brightness:

[0144] Histogram analysis: By analyzing the brightness histogram of the photo, understand the brightness distribution of the image. An ideal photo brightness histogram should show a relatively uniform distribution, avoiding over-bright or over-dark areas. If the histogram is overly concentrated at one end, it indicates an exposure problem in the photo. For example, if the histogram is concentrated in the high-brightness area, it may mean the photo is overexposed; if it is concentrated in the low-brightness area, it may be underexposed.

[0145] Local contrast: Calculate the contrast of different regions of the photo to ensure that key areas such as the tourist's face have sufficient contrast to highlight details. For example, use the Laplacian operator or gradient operator to calculate the local contrast of the image. A higher local contrast means the image details are clearer.

[0146] Color harmony:

[0147] Color histogram matching: Analyze the color histogram of the photo to evaluate whether the distribution and combination of colors are harmonious. The color histogram of the photo can be compared with the histograms of some reference images considered to have harmonious colors, and calculate the similarity between them. The higher the similarity, the more harmonious the color combination of the photo.

[0148] Dominant color and emotional expression: Different colors convey different emotions. The theme of being happy is usually associated with bright and warm colors (such as red and yellow). Analyze the proportion and distribution of the dominant color in the photo to judge whether it helps to express the happy mood. For example, if the warm colors such as red and yellow account for a relatively high proportion in the photo and are distributed around the tourist, it can better set off a happy atmosphere.

[0149] 4. Comprehensive evaluation and ranking

[0150] Weight Setting: Different weights are assigned to each of the above evaluation indicators (such as smile degree, eye vitality, composition, lighting, color, etc.). The weight setting can be optimized based on experience or through machine learning methods. For example, for photos that mainly highlight the expressions of tourists, the weights of smile degree and eye vitality can be relatively high; while for photos that focus on overall aesthetic feeling, the weights of composition and color can be appropriately increased.

[0151] Comprehensive Score Calculation: According to the set weights, each evaluation indicator of each photo is quantitatively scored, and the comprehensive score is calculated. For example, the score of smile degree is multiplied by the corresponding weight, added to the score of eye vitality multiplied by its weight, and then added to the sum of the scores of composition, lighting, color, etc. multiplied by their respective weights, which is the comprehensive score of the photo.

[0152] Sorting and Screening: All photos are sorted according to the comprehensive score, and the photo with the highest score is the photo in the best state under the current evaluation criteria. According to actual needs, several photos with the highest scores can be selected as alternatives.

[0153] Furthermore, the comprehensive score calculation is based on the principle of information entropy, and the weights are determined according to the variation degree of each index data. The greater the variation degree of the data, the more information the index carries, and the higher the weight. The calculation steps are as follows:

[0154] Data Standardization: The original data x ij (i represents the photo serial number, j represents the index serial number) is standardized to obtain y ij , eliminating the influence of dimension.

[0155] Calculating Information Entropy: Calculate the information entropy of the jth index where m is the number of photos.

[0156] Calculating Entropy Weight: The entropy weight of the jth index n is the number of indexes.

[0157] Calculate the comprehensive score, and the formula is as follows:

[0158]

[0159] where x i is the score of each index after standardization; n i is the number of indexes of the ith photo; represents the total number of indexes of all photos participating in the comparison this time.

[0160] Furthermore, the calculation steps of x i are exemplified as follows:

[0161] Min - Max normalization: For example, when calculating the comprehensive score of photos using the entropy weight method, for the "clarity" index, the original values of 10 photos are 20, 30, 40, 50, 60, 15, 25, 35, 45, 55. The minimum value x min = 15, and the maximum value x max = 60. For a certain photo with the original "clarity" value x = 40, after normalizing to [0, 1],

[0162] S3: Obtain the time and geographical location corresponding to the first photo.

[0163] Furthermore, the geographical location and time when the first photo was taken can be recorded through the GPS module or Beidou positioning module of the camera and the network time.

[0164] Furthermore, it is also possible to connect with devices such as the tourist's mobile phone to obtain the geographical location by getting the data of the mobile phone's GPS module.

[0165] Furthermore, the photo information can also include information such as the weather.

[0166] Many video files themselves contain metadata, which may record information such as the shooting time and location. The following tools can be used to view or obtain it:

[0167] MediaInfo: It can analyze the detailed information of various video formats, including data in aspects such as containers, videos, and audios. After installation, open the video file to view the metadata.

[0168] metadata2go: A tool for online viewing of video metadata. Without the need for downloading and installation, upload the video file or enter the video link address to view.

[0169] Wondershare UniConverter: It can be used to edit videos and also view video metadata.

[0170] ivMeta: Specifically used for extracting iPhone video metadata, and can obtain information such as the video shooting date and GPS coordinates.

[0171] S4: Associate the geographical location where the first photo was taken with the relevant geographical locations on the map and mark it on the map.

[0172] Specifically: Mark the first photo successively at the relevant geographical locations on the map, that is, display the first photo taken at the geographical location on the map where the first photo was shot.

[0173] S5: Arrange the marked relevant geographical locations in chronological order of the shooting times of all the first photos on the map to generate a check-in track photo album.

[0174] In the prior art or common general knowledge, taking check-in photos for mementos is carried out through devices such as mobile phones and cameras, and all require manual shooting. This will inevitably lead to most of the image materials being posed, and the check-in locations on the photo album cannot reflect the real feelings during traveling. At the same time, since tourists cannot take photos all the time, some locations worth checking in and happy moments cannot be photographed in time and are thus omitted. Moreover, taking manual photos will make tourists busy finding locations for check-in photos, and most of the time they have no intention of deeply experiencing the scenic spots, which affects the travel experience.

[0175] Therefore, in this application, the cameras in the tourist area are used to automatically collect images of tourists. After analyzing the expressions of tourists in the images, instant photos in a happy, excited or other states worth taking photos for mementos are obtained, and a check-in track photo album is automatically generated based on this. Compared with the manual photo-taking and editing methods using mobile phones and cameras, many accidental happy moments of tourists can be collected, and each happy moment of tourists can be recorded from an objective perspective. Compared with the posed photos taken by cameras or mobile phones, the photos taken by the cameras are closer to the real mood of tourists, and the generated photo album gives tourists a stronger sense of recall experience and will not omit every happy moment. At the same time, the solution of this application can enable tourists not to constantly worry about where it is appropriate to take photos and only focus on check-in, but can let tourists fully immerse themselves in playing in the tourist area without having to worry about taking photos, thus enhancing the depth and level of the travel experience.

[0176] S6: Analyze each of the first photos on the check-in track photo album to obtain the natural environmental conditions of each first photo; obtain the corresponding poems and songs according to the natural environmental conditions and display them in association with the first photos, which can help tourists with basic photo album editing (especially when they can't remember many poems and songs). Specifically, it includes the following steps:

[0177] 1. Based on image recognition and database matching

[0178] Image feature extraction

[0179] Use computer vision technologies such as convolutional neural networks (CNNs) to analyze the photos and extract the key features in the images, such as objects, scenes, colors, textures, etc. For example, identify the main elements such as mountains, waters, people, and flowers in the photos. Fine-tune based on classic models such as AlexNet and VGG to adapt to the feature extraction tasks of different types of photos.

[0180] Establish a poem database

[0181] Collect and organize a large number of poems, songs, and odes, and classify and label them according to dimensions such as theme, emotion, and imagery. For example, classify poems describing mountains and waters into one category and poems expressing joy into another category. Mark each poem with corresponding key information, such as the imagery it contains (e.g., "willow" represents parting), the emotion it expresses (happy, sad, etc.), and the applicable scenarios (farewell, scenery description, etc.).

[0182] Matching process

[0183] Match the extracted photo features with the annotation information in the poem database. For example, if the photo is recognized as a mountain forest scene in autumn, retrieve poems related to autumn scenery and mountain forests from the database. Similarity calculation methods, such as cosine similarity, can be used to measure the matching degree between the photo features and the poem annotation information, and select the poem with the highest similarity as the matching result.

[0184] Furthermore, cosine similarity measures the similarity between two vectors by calculating the cosine value of the angle between them. The closer the value is to 1, the more similar the two vectors are. When measuring the matching degree between photo features and poem annotation information, the photo features and poem annotation information need to be converted into vector form first. Assume the photo feature vector is The poem annotation information vector is The cosine similarity calculation formula is:

[0185]

[0186] Where:

[0187] is the dot product of vectors and The calculation method is to multiply the corresponding dimension elements and then sum them up, that is

[0188] and are the L2 norms (Euclidean norms) of vectors and respectively. The calculation method is the square root of the sum of the squares of the elements in each dimension, that is and

[0189] For example, if there is a photo feature vector The poem annotation information vector Then:

[0190]

[0191] Substitute the above results into the formula to get:

[0192]

[0193] This indicates that the similarity between these two vectors is extremely high.

[0194] Furthermore, the method for extracting the photo feature vector is as follows: The extraction method is as follows:

[0195] 1. Scale-Invariant Feature Transform (SIFT)

[0196] Principle: By detecting key points in the image and calculating the gradient direction and amplitude in the regions around these key points, a feature descriptor that is invariant to image scale, rotation, and illumination changes is generated. It first constructs a Gaussian pyramid of the image, detects extreme points as key points in different scale spaces, then assigns a main direction to each key point, and finally calculates the gradient direction histogram within its neighborhood centered on the key point to generate a 128-dimensional feature vector.

[0197] Application scenarios: Suitable for scenarios such as image matching and object recognition. For example, when stitching panoramic images, SIFT feature vectors are used to find corresponding points between different images.

[0198] 2. Speeded-Up Robust Features (SURF)

[0199] Principle: Based on the determinant of the Hessian matrix to detect interest points in the image, and then generate a feature descriptor by calculating the Haar wavelet response within the neighborhood centered on the interest point. SURF also has good invariance to scale, rotation, and illumination changes, and its calculation speed is faster than SIFT. The dimension of the feature vector it generates is usually 64-dimensional or 128-dimensional.

[0200] Application scenarios: Commonly used in computer vision tasks with high real-time requirements, such as the visual navigation of mobile robots, which can quickly extract image features for scene recognition and positioning.

[0201] 3. Histogram of Oriented Gradients (HOG)

[0202] Principle: The image is divided into small cell units, the direction histogram of the gradient within each unit is calculated, and then these histograms are combined to form a feature vector. HOG features emphasize the edge and shape information of objects in the image and have good stability to geometric and optical deformations of the target. In pedestrian detection, HOG features are widely used to describe the external features of pedestrians by calculating the gradient direction histogram of the pedestrian area in the image.

[0203] Application scenarios: Particularly suitable for object detection tasks, especially for detecting objects with obvious edge features such as humans.

[0204] Furthermore, the method for extracting the poetry annotation information vector is as follows:

[0205] 1. Bag-of-Words (BoW)

[0206] Principle: Consider the poem text as a "bag" without considering the order of words, and only count the number of times each word appears in the poem. First, construct a dictionary containing all the non-repeating words that appear in the poems. Then, for each poem, according to the order of words in the dictionary, count the frequency of each word in that poem to form a vector. For example, for the poem "Before my bed a pool of light, I wonder if it's frost on the ground", the dictionary is {"Before my bed", "bright moon", "light", "wonder if", "on the ground", "frost"}, then the corresponding bag-of-words vector for this poem may be [1, 1, 1, 1, 1, 1].

[0207] 2. TF-IDF (Term Frequency-Inverse Document Frequency)

[0208] Principle: TF (Term Frequency) represents the frequency of a word appearing in a poem, and IDF (Inverse Document Frequency) measures the rarity of a word in the entire poem collection. The TF-IDF value is the product of TF and IDF, and the calculation formula is: TF-IDF i,j = TF i,j × IDF i , where TF i,j is the frequency of word i appearing in the poem, N is the total number of poems in the poem collection, and n i is the number of poems containing word i. The vector obtained through TF-IDF calculation highlights the words that appear frequently in the current poem and rarely appear in other poems, and can better reflect the uniqueness of the poem.

[0209] Advantages: Compared with the bag-of-words model, TF-IDF can, to a certain extent, distinguish the importance of different words to the poem content, reduce the influence of common words, and improve the representativeness of the vector for the poem semantics.

[0210] S7: Display the check-in track album to the tourists, obtain the photo selection information input by the tourists, and regenerate the check-in track album according to the selection information.

[0211] It also includes the following steps:

[0212] S8: Automatically collect the images of the tourists and analyze the images of the tourists to obtain the body movement information of the tourists; if the tourists make specific body movements, take pictures of the tourists to obtain the fifth photo. The analysis of the images of the tourists to obtain the body movement information of the tourists specifically includes the following steps:

[0213] 1. Data collection and preprocessing

[0214] Data collection:

[0215] Multi-scenario collection: In different scenic spots such as squares, trails, and viewing platforms, a large number of images containing tourists are captured through fixed cameras or mobile devices. These images should cover various weather, lighting conditions, and different time periods to ensure data diversity.

[0216] Multi-angle shooting: Tourists are photographed from different angles, including the front, side, back, etc., to comprehensively capture the performance of tourists' body movements from different perspectives.

[0217] Data annotation:

[0218] Define specific actions: Clearly define specific body movements that need to be judged, such as waving, jumping, bending, etc., and formulate detailed annotation rules for each action.

[0219] Manual annotation: Using professional image annotation tools such as LabelImg, trained annotators mark the body key points of tourists (such as head, shoulders, elbows, wrists, hips, knees, ankles, etc.) on the images and annotate whether a specific action is made. For images with specific actions, record the action type and the corresponding key point positions; for images without specific actions, also mark the key point positions as a comparison.

[0220] Data preprocessing:

[0221] Image normalization: Adjust all images to a unified size, such as 224x224 pixels, for subsequent model processing. At the same time, perform normalization processing on the brightness, contrast, color, etc. of the images to reduce the impact of lighting and color differences on the model.

[0222] Divide the dataset: Divide the labeled image dataset into a training set, a validation set, and a test set, usually in a ratio of 7:2:1. The training set is used to train the model, the validation set is used to adjust the model parameters and prevent overfitting, and the test set is used to evaluate the final performance of the model.

[0223] 2. Select a suitable model

[0224] Deep learning-based models:

[0225] Convolutional Neural Network (CNN): Classic CNN models such as ResNet and VGG consist of a network structure of convolutional layers, pooling layers, and fully connected layers, which automatically learn the feature representations in images. When processing image data, CNN can effectively extract the local features of images and has good results for the recognition of body movements. For example, based on a pre-trained ResNet model, it can be fine-tuned according to the task of specific action recognition.

[0226] Pose Estimation-based Model: OpenPose is a model widely used in human pose estimation. It can not only detect the key points of the human body but also determine the connection relationships between key points through Part Affinity Fields (PAFs), thus accurately restoring the human pose. Using the output results of the OpenPose model, it is possible to further analyze whether a tourist makes a specific limb movement. For example, by detecting the positional relationship between the wrist and shoulder key points and the movement trajectory of the arm, it can be judged whether a waving action is made.

[0227] Spatio-temporal Model: For video data, a 3D Convolutional Neural Network (3D-CNN), such as the C3D model, can be used. 3D-CNN can simultaneously process the spatial and temporal dimension information of images. By performing 3D convolution operations on consecutive multiple frames of images, it learns the spatio-temporal feature patterns of limb movements. For example, when judging a jumping action, 3D-CNN can capture the position changes and action postures of the human body in several consecutive frames of images, thus more accurately identifying the action.

[0228] Traditional Machine Learning Models (for auxiliary or comparative use):

[0229] Support Vector Machine (SVM): Based on the extraction of handcrafted features (such as Scale-Invariant Feature Transform SIFT, Histogram of Oriented Gradients HOG, etc.), SVM is used for classification. For example, first, the contour and gradient information of the human body in the image are obtained through the HOG feature extraction method, and then these features are input into the SVM classifier to judge whether a tourist makes a specific action. Although the performance of traditional machine learning models may be inferior to that of deep learning models in complex scenarios, they can be used as comparative experiments to help understand the model performance and the effectiveness of features.

[0230] 3. Model Training and Optimization

[0231] Training Process:

[0232] Parameter Setting: According to the characteristics of the selected model, appropriate training parameters are set, such as learning rate, batch size, number of training epochs, etc. For example, for a CNN-based model, the initial learning rate can be set to 0.001, the batch size to 32, and the number of training epochs to 50.

[0233] Loss Function Selection: Select an appropriate loss function according to the task type. For a binary classification problem (judging whether a specific action is made), the cross-entropy loss function can be used. The cross-entropy loss function can measure the difference between the prediction result and the true label, and the model parameters are optimized by minimizing the loss function.

[0234] Training Execution: Input the training set data into the model and perform training according to the set parameters. During training, the model continuously adjusts its own parameters to minimize the value of the loss function, thereby improving its recognition ability for specific actions.

[0235] Model Optimization:

[0236] Verification and Parameter Tuning: Use the validation set to evaluate the model during training and observe the performance metrics of the model on the validation set (such as accuracy, recall, etc.). According to the validation results, adjust the model's parameters, such as the learning rate, network structure, etc., to prevent overfitting and improve the generalization ability of the model. For example, if it is found that the model has a high accuracy on the training set but a low accuracy on the validation set, it indicates that there may be an overfitting problem, and one can try to reduce the number of network layers or increase the regularization term.

[0237] Data Augmentation: To increase the diversity of data and improve the generalization ability of the model, data augmentation techniques can be used during training. For example, perform operations such as random rotation, flipping, and scaling on images so that the model can learn the limb movement characteristics under different postures and perspectives.

[0238] 4. Model Evaluation and Application

[0239] Model Evaluation:

[0240] Metric Selection: Use the test set to evaluate the trained model. The main evaluation metrics include accuracy, recall, F1 value, etc. Accuracy represents the proportion of the number of samples correctly predicted by the model to the total number of samples; recall represents the proportion of actual positive samples that are correctly predicted as positive samples by the model; the F1 value is the harmonic mean of accuracy and recall, comprehensively reflecting the performance of the model.

[0241] Evaluation Analysis: Analyze the performance of the model under different metrics to understand the advantages and disadvantages of the model. For example, if the accuracy is high but the recall is low, it indicates that the model may have a problem of insufficient recognition of positive samples (samples making specific actions), and it is necessary to further optimize the model's recognition ability for specific actions.

[0242] Practical Application:

[0243] Real-time Monitoring: Deploy the trained model to an actual scenario, such as the monitoring system in a scenic area, analyze the real-time collected images of tourists, and judge in real-time whether tourists make specific limb movements, such as the action of asking for help with taking pictures.

[0244] S9: Obtain the time, geographical location, and corresponding poems and songs of the fifth photo, and display them in association in the check-in track album in the order of shooting time.

[0245] Steps S8 and S9 can enable the function of taking pictures at any time and anywhere in the tourist area without the help of others, without tools, and even without a mobile phone or camera, realizing a seamless and free travel experience (that is, traveling without bringing or bringing fewer tools does not affect the travel experience).

[0246] Embodiment 2

[0247] As Figure 2 shown, a system for improving the tourist attraction experience in a tourist area, which is applied to the method for improving the tourist attraction experience in a tourist area described above, includes:

[0248] A first acquisition module, which is used to automatically acquire images of tourists, analyze the images of tourists, and obtain the emotional state of tourists;

[0249] A first separation module, which is used to separate and obtain the first photos of tourists in the first emotional state;

[0250] A second acquisition module, which is used to obtain the time and geographical location corresponding to the first photos;

[0251] A first annotation module, which is used to associate the geographical location where the first photos are taken with the relevant geographical locations on the map and make annotations on the map;

[0252] A generation module, which is used to arrange the annotated relevant geographical locations in chronological order of the shooting times of all the first photos on the map to generate a check-in track photo album;

[0253] By automatically acquiring images of tourists to generate a check-in track photo album, the photos taken are closer to the real mood of tourists, and there will be no omission of memorable moments. The generated photo album makes the tourists' memory experience stronger. At the same time, tourists don't need to worry about taking pictures and can fully immerse themselves in playing, enhancing the depth and level of the travel experience.

[0254] Embodiment 3

[0255] A computer-readable storage medium, the computer-readable storage medium includes a stored program, wherein, when the program runs, it controls the device where the computer-readable storage medium is located to execute the method for improving the tourist attraction experience in a tourist area described above.

[0256] Embodiment 4

[0257] A processor, the processor is used to run a program, wherein, when the program runs, it executes the method for improving the tourist attraction experience in a tourist area described above.

[0258] The present application provides a method for improving the tourist attraction experience in a tourist area, including: S1: Automatically collecting images of tourists and analyzing the images of tourists to obtain the emotional state of the tourists; S2: Separating and obtaining the first photos of the tourists in the first emotional state; S3: Obtaining the time and geographical location corresponding to the first photos; S4: Associating the geographical location captured in the first photos with the relevant geographical locations on the map and marking them on the map; S5: Arranging the marked relevant geographical locations in chronological order of the shooting times of all the first photos on the map to generate a check-in track photo album; By automatically collecting images of tourists to generate a check-in track photo album, the photos taken are closer to the true mood of the tourists and will not miss any memorable moments. The generated photo album enhances the tourists' memory experience. At the same time, it allows tourists to focus on playing without worrying about taking pictures, improving the depth and level of the tourism experience.

[0259] Those of ordinary skill in the art can realize that the units of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the components of each example have been generally described according to their functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0260] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored, etc.

[0261] In addition, the functional units in each embodiment of the present invention can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0262] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, read-only memories (ROMs), random access memories (RAMs), mobile hard disks, magnetic disks, or optical discs.

[0263] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the various embodiments of the present invention, and they should all be covered within the scope of the claims and the description of the present invention.

Claims

1. A method for improving tourist attraction experience in a tourist area, characterized in that: include: S1: Automatically collect images of tourists and analyze them to obtain their emotional states; S2: separating and obtaining the first photo of the tourist in the first emotional state; S3: Obtaining the time and geographical location corresponding to the first photo; S4: Associating the geographical location taken by the first photo with the relevant geographical location on the map, and marking it on the map; S5: Arrange the marked relevant geographical locations in chronological order according to the shooting time of all the first photos on the map to generate a check-in track album; By automatically collecting images of tourists and generating check-in track albums, the photos taken are closer to the tourists' real moods and will not miss any memorable moments. The albums generated by screening photos with the best emotional states make tourists' memories more vivid. Tourists do not need to worry about taking pictures and can devote themselves to the tour, which enhances the depth and level of the travel experience.

2. The method for improving tourist attraction experience in a tourist area according to claim 1, characterized in that: When analyzing the images of tourists, only part of the images of the video data are collected for emotion analysis; the collection of part of the images adopts a timed collection method.

3. The method for improving tourist attraction experience in a tourist area according to claim 1, characterized in that: The step of separating and obtaining a first photo of a tourist in a first emotional state comprises the following steps: separating and obtaining a plurality of second photos of the tourist in the first emotional state; Find the best photo among the second photos as the third photo; Taking the time corresponding to the third photo as the acquisition reference, re-acquire images of the video data within the third time before and after to obtain a plurality of fourth photos; A plurality of fourth photos are compared and analyzed with the third photo, and the photo in the best state is obtained as the first photo.

4. The method for improving tourist attraction experience in a tourist area according to claim 3, characterized in that: The obtaining of the photo in the best condition refers to calculating a comprehensive score of the photos, and taking the photo with the highest score as the photo in the best condition.

5. The method for improving tourist attraction experience in a tourist area according to claim 4, characterized in that: The comprehensive score calculation includes the following steps: Data standardization, the original data x of each indicator ij After standardization, we can get y ij , to eliminate the dimension effect, where i represents the photo number and j represents the indicator number; Calculate the information entropy. The formula for calculating the information entropy of the jth indicator is as follows: and j =-k∑ m i＝1 p ij lnp ij ; in m is the number of photos; Calculate the entropy weight. The entropy weight calculation formula for the jth indicator is as follows: n is the number of indicators; The comprehensive score is calculated using the following formula: where x i is the standardized score of each indicator; n i is the number of indicators of the i-th photo; Indicates the total number of indicators of all photos involved in this comparison.

6. The method for improving tourist attraction experience in a tourist area according to claim 1, characterized in that: The following steps are also included: S6: Analyze each first photo in the check-in track album to obtain the natural environment conditions of each first photo; Obtaining corresponding poems and songs based on natural environmental conditions and displaying them in association with the first photo can help tourists perform basic photo album editing; S7: Show the check-in track album to the tourist, obtain the album photo selection information input by the tourist, and regenerate the check-in track album according to the selection information.

7. The method for improving tourist attraction experience in a tourist area according to claim 1, characterized in that: The following steps are also included: S8: automatically collecting the image of the tourist and analyzing the image of the tourist to obtain the tourist's body movement information; if the tourist makes a specific body movement, taking a photo of the tourist to obtain a fifth photo; S9: Obtain the time, geographic location, and corresponding poem or song corresponding to the fifth photo, and display them in association in the check-in track album in the order of shooting time.

8. A system for improving the tourist experience in a tourist area, characterized in that: The method for improving the tourist attraction experience in a tourist area as claimed in any one of claims 1 to 7 comprises: The first acquisition module is used to automatically acquire images of tourists and analyze the images of tourists to obtain the emotional state of tourists; A first separation module, which is used to separate and obtain a first photo of the tourist in a first emotional state; A second acquisition module, which is used to obtain the time and geographical location corresponding to the first photo; A first marking module, which is used to associate the geographical location taken by the first photo with the relevant geographical location on the map, and mark it on the map; A generation module, which is used to arrange the marked relevant geographical locations on the map in the order of the shooting time of all the first photos, and generate a check-in track album; By automatically collecting images of tourists and generating check-in track albums, the photos taken are closer to the tourists' real moods and will not miss the moments worth remembering. The generated albums make tourists' memories more powerful. At the same time, tourists do not need to worry about taking pictures and can devote themselves to the tour, which enhances the depth and level of the travel experience.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the method for improving the tourist attraction experience in a tourist area as described in any one of claims 1 to 7.

10. A processor, characterized in that: The processor is used to run a program, wherein the program, when running, executes the method for improving the tourist attraction experience in a tourist area as described in any one of claims 1 to 7.

Images